Prevalence of malocclusions and oral dysfunctions in children with persistent sleep-disordered breathing after adenotonsillectomy in the long term
2020; American Academy of Sleep Medicine; Volume: 16; Issue: 8 Linguagem: Inglês
10.5664/jcsm.8534
ISSN1550-9397
AutoresJulia Cohen-Lévy, Marie‐Claude Quintal, Pierre Rompré, Fernanda R. Almeida, Nelly Huynh,
Tópico(s)Tracheal and airway disorders
ResumoFree AccessScientific InvestigationsPrevalence of malocclusions and oral dysfunctions in children with persistent sleep-disordered breathing after adenotonsillectomy in the long term Julia Cohen-Levy, DDS, Ms, PhD, Marie-Claude Quintal, MD, Pierre Rompré, Fernanda Almeida, DMD, Ms, PhD, Nelly Huynh, PhD Julia Cohen-Levy, DDS, Ms, PhD Orthodontic Clinic, Faculty of Dentistry, Université de Montréal, Canada , Marie-Claude Quintal, MD Ear Nose and Throat Department, Sainte-Justine Pediatric Hospital, Montreal, Canada , Pierre Rompré Statistics Department, Faculty of Dentistry, Université de Montréal, Canada , Fernanda Almeida, DMD, Ms, PhD Faculty of Dentistry, University of British Columbia, Canada , Nelly Huynh, PhD Orthodontic Clinic, Faculty of Dentistry, Université de Montréal, Canada Research Centre, Sainte-Justine Pediatric Hospital, Canada Published Online:August 15, 2020https://doi.org/10.5664/jcsm.8534Cited by:7SectionsAbstractPDFSupplemental Material ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTStudy Objectives:To evaluate the prevalence of craniofacial/orthodontic abnormalities and oral dysfunctions in a population of children with persistent sleep-disordered breathing despite adenotonsillectomy.Methods:Medical charts of 4,000 children with sleep-disordered breathing operated on in a tertiary hospital were retrospectively reviewed. Patients reporting persistent sleep-disordered breathing symptoms were invited to an orthodontic/myofunctional evaluation following the Sleep Clinical Score), followed by a 1-night ambulatory type III sleep study.Results:One hundred nonsyndromic symptomatic patients were examined (mean age 8.8 ± 3.5 years), from 1 to 12 years after surgery (mean 4.6 ± 3.1 years); 24% were overweight/obese; 69 had a sleep study. Although prevalent, oronasal abnormalities and malocclusions were not specifically associated with pathological sleep parameters (cartilage hypotonia 18%, septal deviation 5%, short lingual frenulum 40%). Malocclusions were associated with a higher respiratory event index in children under 8 years only, whereas an impaired nasal dilator reflex and tongue immaturity were associated with an increased obstructive respiratory event index in all patients (1.72 ± 2.29 vs 0.72 ± 1.22 events/h, P = .011) and Respiratory Event Index, respectively (3.63 ± 3.63 vs 1.19 ± 1.19 events/h). Male sex, phenotype, nasal obstruction, oral breathing, and young age at surgery (< 3 years) were significantly related to higher respiratory event index. Using the Sleep Clinical Score > 6.5 cut-off, patients with persistent sleep apnea were significantly distinct from chronic snoring (2.72 ± 2.67 vs 0.58 ± 0.55, P < .01).Conclusions:Oronasal anatomical and functional abnormalities were quite prevalent and various in persistent sleep-disordered breathing after adenotonsillectomy. Nasal disuse and tongue motor immaturity were associated with a higher obstructive respiratory event index in the long term, whereas craniofacial risk factors might have a more pronounced impact at younger age.Citation:Cohen-Levy J, Quintal M-C, Rompré P, Almeida F, Huynh N. Prevalence of malocclusions and oral dysfunctions in children with persistent sleep-disordered breathing after adenotonsillectomy in the long term. J Clin Sleep Med. 2020;16(8):1357–1368.BRIEF SUMMARYCurrent Knowledge/Study Rationale: Long-term studies of adenotonsillectomy or sole adenoidectomy suggest an incomplete resolution of sleep-disordered breathing in certain populations or its possible relapse with time. The aim of this study was to evaluate the prevalence of craniofacial, oronasal abnormalities, and dysfunctions in nonsyndromic children and adolescents with persistent sleep-disordered breathing after surgery.Study Impact: In our sample of mild to moderate persistent sleep-disordered breathing, several clinical phenotypes were found, including a prevalence of 24% in children and adolescents who were overweight/obese. Oral breathing impaired nasal dilator reflex and tongue motor immaturity were more associated with increased apnea-hypopnea indexes than were craniofacial features or malocclusions. The Sleep Clinical Record (when score > 6.5) could adequately discriminate persistent obstructive events from persistent chronic snoring.INTRODUCTIONSleep-disordered breathing (SDB), a spectrum of disorders affecting both pediatric and adult populations, arises from repetitive episodes of partial or complete obstruction of the upper airways during sleep. SDB has various degrees of severity,1 starting with primary snoring, affecting approximately 12% of children, to obstructive sleep apnea (OSA), which affects from 1.2 to 5.7%.2 Left untreated, OSA has been related to significant morbidity in children: growth failure, neurocognitive and behavioral abnormalities, reduced academic performance, and also cardiovascular effects.SDB conditions share a common pathophysiology, ie, the narrowing of upper airways, resulting from an increased soft tissue content, a volumetric reduction of the facial skeletal frame, an alteration of its neuromuscular tone, or a combination of those factors.In children, the most described cause of upper airway obstruction is the hypertrophy of adenoids and palatal tonsils3 (part of Waldeyer's lymphoid ring), potentially aggravated by a nasal obstruction and associated with mild abnormalities of the facial skeleton, such as a narrow maxilla and/or a retruded, steep mandible.4 The asynchronism of lymphoid growth velocity compared with skeletal growth5 explains the peak prevalence of SDB, clinically observed between 3 and 5 years of age6, but some patients might be affected with obesity, similar to the most frequent type of OSA in adults. Syndromic patients constitute another phenotype, associated with major alterations of craniofacial growth or neuromuscular tone and they display early and complex forms of OSA.In children with symptoms, polysomnography (PSG) is the gold standard method to score respiratory events. There is agreement on the definition of pathological indices in children, with a pathological limit of obstructive apnea-hypopnea index > 1.5 events/h: an index < 5 defines a mild OSA, an index between 5 and 10 is a moderate OSA, and an index exceeding 10 is a severe OSA.For patients whose treatment cannot be postponed, especially those with a high apnea-hypopnea index, adenotonsillectomy (A&T) is the first-line treatment of OSA. The CHAT (Childhood Adenotonsillectomy Trial) study prospectively evaluated OSA outcomes after A&T and reported an overall success of 79%, defined as an apnea-hypopnea index < 2 events/h and an obstructive apnea index < 1 events/h. Nevertheless, several studies report the persistence of SDB in 20 to 40% of patients after A&T,7,8 the success rate being significantly worse when residual OSA is defined as an apnea-hypopnea index >1 events/h on postoperative PSG rather than reported symptoms.9Data on long-term outcomes of A&T and sole adenoidectomy are scarce,10,11 but it has been suggested that there is an incomplete resolution of SDB in certain populations, such as patients with craniofacial abnormalities, especially syndromic (such as in Down syndrome12), patients with obesity, or children with underlying asthma or allergic rhinitis.13 SDB could also possibly relapse with time.14Therefore, the aim of the present study was to assess the long-term prevalence of persistent SDB symptoms following adenoid and/or tonsil surgery and to evaluate the contribution of various risk factors, such as obesity, craniofacial nonsyndromic features4 (transverse deficiency of the jaws, maxillary and/or mandibular sagittal deficiencies—retrognathia—and/or excessive or deficient vertical growth, with associated malocclusions) and oronasal dysfunctions, that can persist and worsen beyond early childhood.METHODSStudy designThis is a prospective observational study. The study protocol was approved by the Ethical Committee of St Justine Hospital, Montreal, Canada (# 2016-1033), in accordance with the 1964 Helsinki declaration and its later amendments. Consent was obtained from the children's legal guardians, and patients themselves when able to understand and sign the informed consent forms, for each section of the study.ParticipantsParticipants were recruited from the Ear Nose and Throat unit of a pediatric tertiary hospital, who had previously undergone surgery of adenoids and/or tonsils between January 2000 and March 2016, with the necessary report of chronic snoring or witnessed apneas in their preoperative clinical charts. They were invited to participate in this study, which included 3 separate visits (V1–V3). During V1, participants completed the hierarchic severity clinical scale (HSCS)15,16 to assess self-reported persistent SDB symptoms; these results were previously published.17 Participants who reported chronic snoring (HSCS > 0) or who scored positive for suspected OSA (HSCS > 2.72) in absence of craniofacial syndrome were invited to complete the second (V2) and third (V3) study visits (see Figure 1). V2 consisted of a physical evaluation to assess craniofacial morphology, oronasal function, and body mass index (BMI); V3 consisted of a home sleep apnea test (HSAT). Participants and their parents could choose to complete all or only 1 of these study visits. All study data were collected and managed using a secure electronic data capture tool for research, the RedCap platform (Research Electronic Data Capture), hosted at the Université de Montréal.Figure 1.: Flowchart.A&E = adenotonsillectomy, HSCS = hierarchic severity clinical scale, SDB = sleep-disordered breathing.Download FigurePhysical examinationThe craniofacial morphology and oronasal functional assessments at V2 were completed by a single trained orthodontist (JCL). Results from the physical examination were used to calculate the Sleep Clinical Record total score. This tool was developed by Villa et al18 to identify children with OSA and validated for persistent OSA after treatment19 (see supplemental material) and includes the clinical history and physical examination of each participant.Physical examination comprised the nose (septal deviation, Glatzel mirror test, sniff test, Gudin's reflex20, Rosenthal's test), tongue (relative volume, frenulum21, protraction/elevation tests), oropharynx (Friedman palate and tonsils scales), craniofacial proportions, and occlusion (see supplemental material). Children's height (in cm) and body weight (kg) were measured on an electronic scale (Seca column scale).The BMI was calculated as the body weight divided by the squared height (kg/m2), BMI z-scores (ie, BMI measured in terms of standard deviations from the mean according to age and sex) were calculated using the WHO Anthroplus software (available at https://www.who.int/growthref/tools/en/; see supplemental material). Obesity was defined by BMI z-score of > +2 standard deviations. Clinical history comprised parental report of Attention Deficit Hyperactivity Disorder or other neurological symptom, daytime somnolence, headache, frequent awakenings or agitated sleep, enuresis, and medication.Home sleep apnea testThe HSAT was a 1-night polygraph study done with a type 3 ambulatory device, the NOX T-3 Sleep Monitor (Nox Medical, Reykjavik, Iceland). The system comprises a microphone, a wireless oximeter/cardiac pulse meter, thoracic and abdominal belts, a nasal canula, and oronasal thermistor. This system demonstrated very good measurement agreement compared with in-laboratory PSG and a high degree of sensitivity for detecting even mild OSA.22 Respiratory events were manually scored by an external sleep laboratory ( Sleepstrategies.com, Ottawa, Canada) following the guidelines of the American Academy of Sleep Medicine and indexes calculated accordingly:23 REI (number of respiratory events/monitoring time), obstructive REI (OREI number of obstructive respiratory events/monitoring time), and oxygen desaturation index (ODI; number of 3% desaturation/monitoring time events).Statistical analysisAll results were presented as mean ± standard deviation for continuous variables and as percentage (%) for nominal variables. Statistical analyses were carried out using IBM-SPSS Statistics for Windows (Version 24) for descriptive statistics, parametric (t test), and nonparametric tests (Mann-Whitney U test, Kruskal-Wallis test). The null hypothesis was rejected at P < .05.RESULTSDemographicsOverall, persistent SDB symptoms were reported in 448 participants among those who completed V121 (Figure 1). They were invited to continue onto V2 and V3 between April 2016 and May 2017. However, 70 participants reported spontaneous resolution in the meantime (60 from chronic snoring group, 10 from suspected OSA group). Mild SDB was suspected in 301 children (HSCS > 0 with chronic snoring, 25.59%), whereas 77 (6.54%) had an HSCS > 2.72, suggesting persistent OSA. Among 378 candidates for V2 and V3, 43 were excluded because of a syndromic or compromised medical condition, 152 declined to participate, and 13 missed their scheduled appointments. Thus, 100 consecutive participants with reported persistent SDB symptoms who were representative of the entire cohort underwent the physical examination (V2, see Table 1). Finally, 69 participants completed the HSAT (V3), since 17 declined and 14 refused to repeat their HSAT after it failed initially. For 2 participants, sleep study results were taken from an in-hospital PSG study done at the same study center and within 3 months of V1.Table 1. Characteristics of the surgical cohort, questionnaire, and examined children samples.Male SexAge of Surgery (years)Age at Survey (years)Years from SurgeryDatabase: surgical SDB cohort n = 4,00057.7%4.23 ± 2.1 (0–18)10.57 ± 3.7 (2–19)6.30 ± 3.4 (1–17)n = 2,306V1 Questionnaire Sample n = 1,17656.3%4.34 ± 2.2 (0–18)9.54 ± 3.7 (1.9-19.5)5.19 ± 3.4 (1–16)n = 663V2 Examined Symptomatic Children n = 10063%4.09 ± 2.2 (0–12)8.78 ± 3.5 (2.5-17.5)4.64 ± 3.1 (1–15)n = 63V3 Sleep study n = 6961.4%4.11 ± 2.3 (1–12)9.27 ± 3.6 (4–17)4.11 ± 2.3 (1–15)n = 43SDB = sleep-disordered breathing.General characteristics of the participants at V2 and V3 are described in Table 1. These visits were done between 1 and 15 years after surgery. The study sample was predominantly male, between 63 and 64.4%. Male patients had been operated significantly younger than girls (3.63 ± 1.84 vs 4.92 ± 2.67 years, P = .02). Patients were from various origins (Table 2): 53% of white European origin, 36% from black Africa or Haiti, 13% from North Africa, 5% from other regions. Ethnicity did not influence respiratory events (P = .187 Kruskal-Wallis). Still, children of black African origin tended to be operated at younger age compared with non-African children (3.50 ± 1.79 vs 4.45 ± 2.43 years, P = 0.051). Mean BMI z-score was 1.23 ± 1.51 (range: −1.71 to 5.94), whereas mean waist-to-height ratio was 0.53 ± 0.22 (range: 0.38 to 1.71).Table 2. General anthropometric and surgical factors tested in the symptomatic sample.Studied FactorsFrequency (V2 n = 100)REI (V3 n = 69)P Value and TestSexP = 0.042 Mann-Whitney U Men63% (n = 63)2.12 ± 2.55 (60.9%, n = 42) Women37% (n = 37)1.19 ± 1.65 (39.1%, n = 27)Ethnicity*Mann-Whitney U tests:/other non-East African P = .638/other non-North African P = .064/other non-European P = .229Kruskal-Wallis test:P = .187 East African/Haiti36% (n = 36)2.29 ± 2.67 (30.5%, n = 21) North African13% (n = 13)0.84 ± 2.40 (17.4%, n = 12) European53% (n = 53)1.80 ± 2.24 (55.1% n = 38) Other5% (n = 5)-Type of surgery Sole adenoidectomy21% (n = 21)1.89 ± 1.89 (20.3% n = 14)/ A&T P = .160/ sole adenoidectomy P = .02/A&T P = .07 (Kruskal-Wallis test) Tonsillectomy (previous adenoidectomy)13% (n = 13)1.11 ± 0.70 (10.1% n = 7) Adenotonsillectomy66% (n = 66)1.91 ± 2.45 (69.6% n = 48)Age at surgery < 3 years of age26% (n = 26)2.93 ± 3.18 (27.6% n = 19)P = .007 t test (bilat) vs ≥ 3 years74% (n = 74)1.31 ± 1.64 (72.4% n = 50) > 7 years of age8% (n = 8)1.25 ± 1.82 (11.6% n = 8)P = .504 t test (bilat) vs ≤ 7 years of age92% (n = 92)1.83 ± 2.33 (88.4% n = 61)BMI z-score (long term after surgery) < 2 SD74% (n = 74)1.82 ± 2.65 (65.2% n = 45)P = .837 t test (bilat) ≥ 2 SD26% (n = 26)1.70 ± 1.42 (34.8% n = 24)WtHR (follow-up) ≥ 0.5 (weight dependent risk)48% (n = 48)1.79 ± 1.67 (50.7% n = 35)P = .898 t test (bilat) < 0.5 (low risk)52% (n = 52)1.72 ± 2.78 (49.3% n = 34)HSCSP = .802 t test (bilat) ≥ 2.7230% (n = 30)1.85 ± 1.90 (34.8% n = 24) < 2.7270% (n = 70)1.71 ± 2.46 (65.2% n = 45)Sleep Clinical ScoreP < .001 t test (bilat) ≥ 6.556% (n = 56)2.72 ± 2.67 (55.1% n = 38) < 6.544% (n = 44)0.58 ± 0.55 (44.9% n = 31)P values appear in bold when < .05. *Total exceeds 100 as some children had mixed origins. A&T = adenotonsillectomy, HSCS = hierarchic severity clinical scale, REI = respiratory event index, WtHR = waist-to-height ratio.SleepA subgroup of 69 participants with reported SDB symptoms completed a level 3 HSAT. Results showed 32 participants (46.37%) with persistent OSA (REI > 1), with a mean REI of 3.28 ± 2.60 events/h (range 1.1–12.8 events/h), mean OREI of 2.22 ± 2.16 events/h, and mean ODI of 3.87 ± 2.36 events/h. Their mean age was 9.06 ± 3.79 years, and they were evaluated 5.66 ± 3.84 years after surgery. Male sex was significantly associated with higher REI and ODI (P = .04 and P = .03, respectively).Five patients (7%) had moderate OSA (REI > 5), and only 1 had severe OSA (REI > 10). They were not under any medication and had a variety of phenotypes and ages. These patients with moderate to severe OSA were effectively screened with SCS (all had an SCS score > 6.5) but only 2 of 5 had a positive HSCS score (> 2.72).In the full sample (N = 100), 66% had had an A&T, 21% a sole adenoidectomy, and 7% had had a tonsillectomy as a second surgery (previous adenoidectomy in early childhood). In the HSAT subgroup, revision surgery had significantly reduced REI when compared to A&T and sole adenoidectomy (Kruskal-Wallis test with Bonferroni correction, P = .07 and .02, respectively), but REI was not significantly different in the long term when comparing adenoidectomy with A&T. Children who had been operated before 3 years of age (n = 19 of 69) had significantly more abnormal respiratory events than patients operated at a later age (REI of 2.93 ± 3.18 vs 1.31 ± 1.64 years); 8 children were operated after 7 years of age and they did not have more severe REI than the rest of the sample.Obesity affected 24–26% (26 of 100 in V2, 17 of 69 in V3): 15 were obese (BMI z-score comprised between 2 and 3 standard deviations; mean REI = 1.77 ± 1.55), 11 were severely obese (BMI z-score > 3 standard deviations, mean REI = 1.59 ± 1.25 for 9 patients who completed the HSAT).Obese participants did not have more severe REI than lean patients in this sample when comparing both BMI z-score (1.82 ± 2.65 for < 2 standard deviations vs 1.70 ± 1.42 for ≥ 2 standard deviations) and waist circumference to height ratios (1.72 ± 2.78 for waist-to-height ratio < 0.5 vs 1.79 ± 1.67 for waist-to-height ≥ 0.5).Craniofacial, oronasal anatomical and functional factorsChildren from the full sample showed long/dolichofacial faces in 45%; 39% had a labial incompetence with lip strain (Table 3). Almost 3 of 4 children had a convex profile (74%), 56% had an increased facial height, and 53% a malocclusion: posterior crossbite 13%, anterior crossbite 10%, Angle's class II 31%, class III 5%, increased overjet 52%, dental crowding 61%; 8% had had an orthodontic treatment. None of these craniofacial features was associated with differences in REI. A narrow palate (42% of subjects) tended to be associated with higher REI (P = .05; Table 4).Table 3. Craniofacial factors tested in the symptomatic sample.Craniofacial FactorsFrequency (V2 n = 100)REI (V3 n = 69)P Value and TestFacial type (Vertical) Mesofacial (normal)38% (n = 38)2.03 ± 2.08 (31.9%, n = 22)P = .137 Kruskal-Wallis test Brachyfacial (short face)17% (n = 17)1.94 ± 1.60 (20.3%, n = 14) Dolichofacial (long face)45% (n = 45)1.50 ± 2.63 (47.8%, n = 33) Increased facial height56% (n = 56)1.93 ± 2.76 (58.0%, n = 40)P = .957 Mann-Whitney U vs Normal/Reduced height44% (n = 44)2.40 ± 1.79 (42.0%, n = 29) Labial incompetence39% (n = 39)1.82 ± 3.02 (37.7%, n = 26)P = .336 Mann-Whitney U vs labial competence61% (n = 61)1.71 ± 1.70 (62.3%, n = 43)PhenotypeP = .023 Mann-Whitney U Positive (adenoid or obese)62% (n = 62)2.10 ± 2.48 (68.1%, n = 47) Negative38% (n = 38)1.03 ± 1.54 (31.9%, n = 22)Convexity Straight profile20% (n = 20)2.27 ± 2.26 (17.4%, n = 12)P = .129 Kruskal-Wallis test Convex profile74% (n = 74)1.74 ± 2.33 (76.8%, n = 53) Concave profile6% (n = 6)0.42 ± 0.15 (5.8%, n = 4)Maxilla position Normal65% (n = 65)1.73 ± 2.30 (66.7%, n = 46)P = .154 Kruskal-Wallis test Retruded14% (n = 14)0.80 ± 0.59 (17.4%, n = 12) Protruded21% (n = 21)2.90 ± 2.88 (15.9%, n = 11) Narrow palate42% (n = 42)2.24 ± 2.67 (40.6%, n = 28)P = .054 Mann-Whitney U vs Normal-shape palate58% (n = 58)1.43 ± 1.91 (59.4%, n = 41)Mandible position Normal25% (n = 25)2.96 ± 2.74 (26.1%, n = 18)P = .080 Kruskal-Wallis test Retruded67% (n = 67)1.36 ± 1.99 (69.6%, n = 48) Protruded8% (n = 8)0.97 ± 0.91 (4.3%, n = 3)Dentition Primary18% (n = 18)2.85 ± 1.53 (15.9%, n = 11)P = .281 Kruskal-Wallis test Mixed60% (n = 60)1.53 ± 2.18 (56.5%, n = 39) Permanent22% (n = 22)1.60 ± 2.00 (27.5%, n = 19) Malocclusion (any)53% (n = 53)1.94 ± 2.6 (52.2%, n = 36)P = .750 Mann-Whitney U vs Normal occlusion47% (n = 47)1.56 ± 1.9 (47.8%, n = 33) Posterior crossbite*13% (n = 13)1.07 ± 0.68 (8.7%, n = 6)P = .183 Mann-Whitney U Anterior crossbite*10% (n = 10)0.79 ± 0.63 (13.0%, n = 9)P = .909 Mann-Whitney U Open bite*8% (n = 8)0.98 ± 1.18 (7.2%, n = 5)P = .258 Mann-Whitney U Deep bite* (> 35%)45% (n = 45)1.68 ± 1.88 (46.4%, n = 32)P = .926 Mann-Whitney U Increased overjet*(> 3 mm)52% (n = 52)1.82 ± 2.62 (50.7%, n = 35)P = .987 Mann-Whitney U Normal or reduced overjet48% (n = 48)1.73 ± 1.85 (47.8%, n = 33) Angle's classification* Class I (normal)64% (n = 64)1.80 ± 2.79 (65.2%, n = 45)Kruskal-Wallis test: P = .894 Class II31% (n = 31)1.74 ± 2.08 (29.0%, n = 20) Class III5% (n = 5)1.32 ± 1.85 (5.8%, n = 4) Dental crowding61% (n = 61)1.68 ± 2.27 (68.1%, n = 47)P = .624 Mann-Whitney U vs no crowding/spacing39% (n = 39)1.93 ± 2.31 (31.9%, n = 22).P values appear in bold when < .05. *Total exceeds 100% as some children have combined malocclusions. REI = respiratory events index.Table 4. Nasal, oral, and oropharyngeal factors tested in the symptomatic sample (underlined items are from SCS scoring)Nasal & Oropharyngeal factors (Sleep Clinical Record)Frequency (n = 100)REI (n = 69)P Value, TestOral breathing62% (n = 62)2.16 ± 2.6 (76.8%, n = 53)P = .017 Mann-Whitney UNasal breathing38% (n = 38)1.17 ± 1.50 (23.2% n = 16)Nasal cartilage hypotonia18% (n = 18)1.35 ± 1.25 (20.3%, n = 14)P = .939 Mann-Whitney Uvs no collapsus on inspiration82% (n = 82)1.87 ± 2.48 (79.7%, n = 55)Deviated nasal septum5% (n = 5)0.60 ± 0.20 (4.3%, n = 3)P = .387 Mann-Whitney UNasal obstruction32% (n = 32)2.51 ± 2.42 (30.4%, n = 21)P = .017 Mann-Whitney UNo obstruction68% (n = 68)1.43 ± 2.14 (69.6%, n = 48)Palatal tonsils Score III or IV9% (n = 9)1.13 ± 0.63 (10.1%, n = 7)P = .424 Mann-Whitney U Score 0 (removed), I, II91% (n = 91)1.83 ± 2.38 (89.1%, n = 62)Palate positionP = .059 Mann-Whitney U Score III or IV58% (n = 58)1.51 ± 2.4 (55.1%, n = 38) Score I or II42% (n = 42)2.06 ± 2.1 (44.9%, n = 31)Tongue Short frenulum41% (n = 41)2.25 ± 2.78 (40.6%, n = 28)P = .281 Mann-Whitney U vs normal tongue motility59% (n = 59)1.42 ± 1.80 (59.4%, n = 41) Relative macroglossia38% (n = 38)2.14 ± 2.66 (29.0%, n = 24)P = .126 Mann-Whitney U vs normal tongue volume62% (n = 62)1.56 ± 2.03 (65.2%, n = 45) Motor immaturity (tasks)26% (n = 26)3.63 ± 3.63 (23.2%, n = 16)P < .001 Mann-Whitney U vs ability to elevate tongue74% (n = 74)1.19 ± 1.19 (76.8%, n = 53) Atypical swallowing54% (n = 54)2.08 ± 2.60 (52.2% n = 36)P = .327 Mann-Whitney U vs mature swallowing46% (n = 46)1.52 ± 1.97 (47.8% n = 33)Other factors Parafunctionals habits (> 1)51% (n = 51)1.58 ± 1.95(50.7%, n = 35)P = .656 Mann-Whitney U vs none49% (n = 49)1.94 ± 2.57 (49.3%, n = 34) Reported sleep bruxism27% (n = 27)1.94 ± 1.95 (29.0%, n = 20)P = .643 Mann-Whitney U No reported sleep bruxism73% (n = 73)1.69 ± 1.95 (71.0%, n = 49) Nasal corticoid treatment9% (n = 9)1.32 ± 1.36 (13.0%, n = 9)P = .986 Mann-Whitney U No medical treatment91% (n = 91)1.82 ± 2.38 (87.0%, n = 60)P values appear in bold when < .05. REI = respiratory event index.With children of 8 years of age or less (n = 34) from the HSAT subgroup, the OREI was significantly higher when presenting with at least 1 dental malocclusion (2.19 ± 2.65 events/h, n = 19) compared with normal occlusion (0.63 ± 1.04 events/h, N = 15) but not for a specific malocclusion type.SDB phenotypes according to SCS scoring (including patients with adenoid faces and obesity) were significantly associated with REI (P < .05). Subjects with convex profiles had been operated on significantly younger (3.63 ± 1.84 vs 4.92 ± 2.67 years, P = .02).Oronasal abnormalities were overrepresented: cartilage hypotonia was found in 18 children (all white), with spontaneous nasal collapse on forced inspiration in 13 children (unilateral 4, bilateral 9, see Figure 2A); septal deviation was found in 5 patients. Nasal obstruction (32%) and oral breathing mode (62%) were both significantly associated with a higher REI. Gudin's and Rosenthal tests were failed in 55% and 44% respectively, a failed Gudin's test being significantly associated with higher OREI (P = .011, Mann-Whitney U test, see Figure 3A).Figure 2.: Photographs depicting cartilage hypotonia and short lingual frenulum.(A) Cartilage hypotonia (spontaneous unilateral or bilateral collapse during forced inspiration) was overrepresented in this symptomatic population, but was not related to postoperative sleep parameters nor hierarchic severity clinical scale. (B) Short lingual frenulum (limited tongue elevation in relation to maximal aperture of 60% or less).Download FigureFigure 3.: Obstructive respiratory events index, Gudin's nasal reflex, and tongue immaturity.(A) Obstructive respiratory events index (OREI) per total sleep time (TST) according Gudin's test (median). (B) OREI index per TST according tongue elevation test (median).Download FigureA short lingual frenulum was found in 40% of patients (Figure 2B), relative macroglossia in 38%, and tongue motor immaturity in 24% (patients were unable or did not understand how to elevate the tip of the tongue against the palate when asked). This tongue inability/motor imprecision was significantly associated with increased REI, OREI and ODI (Mann-Whitney U test P < .001, P < .01, and P = .08 respectively, Figure 3B).DISCUSSIONThe American Academy of Pediatrics suggests that children with SDB "should be reevaluated postoperatively to determine whether further treatment is required…. and objective testing should be performed in patients who are high risk or have persistent symptoms/signs of OSA after therapy."2 With the limitation of the HSCS questionnaire survey, adenoidectomy and A&T seemed to be effective in the resolution of SDB symptoms in the long term in about 60% of participants of this study. Mild SDB persistence was reported in 25%, and OSA was suspected in about 7% of nonsyndromic children. When objectively analyzed with a HSAT in a subset of patients, mild to moderate OSA was found in half of examined symptomatic patients: 32 of 69 children had an REI > 1 (46.37%), with a mean REI of 3.28 ± 2.60 events/h (1.1–12.8).A younger age at surgery (under 3 years) was associated with a higher REI in the long term, confirming the findings from Thadikonda et al24 on adenoidectomy alone and Mitchell et al25 on A&T. On the other hand, surgery performed after 7 years of age did not affect REI, contrary to the results of Bhattacharjee et al.10 They had performed a large retrospective study of 578 children who underwent A&T and reported that, when assessed with PSG with a strict criterion of apnea-hypopnea index < 1 event/h, children who were operated on over the age of 7 years were more likely to have persistent OSA.This different outcome might be related to both timing and sleep study method. We chose an HSAT that is less accurate in the pediatric population than for adolescents and adults.26 We also performed the recording a long time after surgery and not in the 6 months to 1 year following surgery; as the maximum peak of lymphoid tissue growth is about age 8 years, residual disease, originating from residual lymphoid tissue, is then more likely to be seen when the study is performed at age 7 or 8 years.Various craniofacial abnormalities were described in preschool children as predisposing factors for SDB, including decreased mandibular and maxillary lengths, skeletal retrusion, and increased lower facial height.27In this study, we could not establish any direct relationship between abnormal respiratory events during sleep and specific malocclusions, such as crossbite, deep bite, or class II occlusion, whose prevalence in this sample was not significantly different from that in the general population.This finding was a surprise, as a strong association is depicted in the literature. In the PANIC study,28 conducted on 491 Finnish children 6–8 years of age, children with crossbite had a 3.3 times higher risk of having SDB (based on a questionnaire) than those without crossbite, and children with a convex facial profile had a 2.6 times higher risk of having SDB than those with a normal facial profile. A recent Italian study,29 comparing a group of 139 children with OSA to a control group of 137 children (range 2–10 years, all diagnosed with PSG), described several orthodontic factors ind
Referência(s)