Tongue strength, hard palate morphology and adenotonsillar hypertrophy in schoolchildren with early mixed dentition: a cross-sectional study

Abstract

Background

Adenotonsillar hypertrophy (ATH) relates to airway obstruction and orofacial changes in children. Whether adenoid hypertrophy (AH), palatine tonsil hypertrophy (PTH), and combined ATH were associated with distinct patterns of hard palate morphology and tongue function in the early mixed dentition remains unclear.

Methods

This cross-sectional study examined 420 schoolchildren aged 6–8 years in central Vietnam. AH was defined by the Fujioka adenoid–nasopharynx ratio (ANR > 0.6) and PTH by Brodsky grades 3–4; tongue strength was measured with the Iowa Oral Performance Instrument, and hard palate morphology by digital dental-cast analysis. Multivariable regression (adjusted for age, sex, BMI, lip-seal at rest, and school location) analyzed individual measures; modified Poisson regression analyzed a co-occurring phenotype (structural alteration plus low tongue strength). Univariable comparisons applied Benjamini–Hochberg false-discovery-rate (FDR) correction.

Results

Enlarged lymphoid tissue was present in 65.2% of children (AH only 19.3%, PTH only 32.6%, combined ATH 13.3%); tapered arch (31.0%), high-arch palate (20.0%), and low tongue strength (35.0%) were common. Nominally, AH was associated with a deeper, shorter palate, and PTH with a narrower tapered arch and lower tongue strength; but none survived FDR correction (smallest q = 0.16). In adjusted models, combined ATH was associated with a smaller anterior maxillary angle (β = −3.76°; 95% CI −6.22 to −1.29; p = 0.003) and palatal vault module (β = −0.54; −1.02 to −0.05; p = 0.032), each decreasing with adenotonsillar burden (both trends p ≤ 0.020). The co-occurring phenotype (77 children; 18.3%) became more common with burden (PTH only aPR 1.77, p = 0.040; combined ATH aPR 1.87, p = 0.048; trend aPR 1.38 per level, p = 0.020).

Conclusions

Adenotonsillar hypertrophy was common in schoolchildren with mixed dentition. AH and PTH showed distinct patterns of palate and tongue alteration, although these did not survive FDR correction and remain exploratory. Combined ATH and greater adenotonsillar burden were consistently associated with maxillary arch narrowing; the co-occurrence alteration was also more frequent with higher burden. These findings support including palatal morphology and tongue function in orofacial screening during the mixed dentition, pending longitudinal confirmation.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12887-026-07186-7.

Introduction

Lingual, palatine, pharyngeal, and tubal tonsils form the Waldeyer’s ring, offering localized immune protection. The palatine tonsil is the main site for lymphatic tissue accumulation, while the adenoid, or pharyngeal tonsil, is situated in the superior and posterior nasopharyngeal airway. The two tonsils play a crucial role in immune responses against foreign antigens, whose inhalation or ingesting can cause the tonsils to become inflamed and enlarged, leading to conditions such as palatine tonsil hypertrophy (PTH), adenoid hypertrophy (AH) or both (adenotonsillar hypertrophy, ATH). A 2018 systematic review and meta-analysis of 764 studies by Pereira et al. concluded that AH prevalence among the general population was 34%, with even higher rates, ranging from 42% to 70%, reported in specific groups such as those referred to ENT clinics or diagnosed with sleep apnea [1]. The prevalence of PTH was found to be 11% in school children [2] or as high as 66.3% in malocclusion children [3].

Hypertrophy of the pharyngeal or palatine tonsils might create respiratory obstruction, especially in the nasopharynx and oropharynx, and is a common cause of upper airway obstruction [1]. Although previous studies have explored the relationship between obstructed upper airways and facial growth, this remains a topic of debate [4]. Thus, Linder-Aronson suggests that nasopharyngeal obstruction promotes mouth breathing and low tongue posture to compensate for increased nasal airflow resistance, potentially leading to various maxillary underdevelopment issues such as transverse constriction and hard palate malformation [5]. Yet, Perez’s study on the influence of tonsil grade on the morphometry of dental arches in 6–12-year-old Chileans found no significant correlation between the maxillary and mandibular dental arches and tonsil grade [6]. Regarding oral motor function, it has been hypothesized that ATH creates a mechanical obstacle that leads to orofacial structural and muscular impairment, including reduced tongue strength [7]. Breathing patterns have been shown to impact tongue pressure development and to interfere with the development of stomatognathic structures [8], implicating muscle impairments [7], In turn, orofacial muscle strength exacerbates the obstructive effects of ATH. Despite its possible interplay, the relationship between ATH and oral muscle strength has not been adequately addressed in the literature except in rare studies [9] or those that focus on a specific population, such as, children with oral breathing [7]. Previous studies have been inconsistent due to small sample sizes, a wide age range that fails to account for physiological changes in adenoid and tonsil size [10] or a population that is predominantly hospital-based and thus unrepresentative of the general population [9, 11, 12]. Thus, further study is needed to investigate links between craniofacial characteristics and oral motor dysfunction on the one hand and different locations of hypertrophied pharyngeal lymphoid tissues on the other.

Moreover, previous research often analyzed AH and PTH together, neglecting to differentiate their effects at the individual level. Huang et al. noted that different combinations of enlarged pharyngeal lymphoid tissue correspond to distinct craniofacial subtypes and highlighted the importance of distinguishing between enlargement of adenoids and of palatine tonsils [11]. Studies comparing the morphological patterns differentiated by etiological factor (i.e., AH versus ATH) are either rare in the literature [4] or do not agree on the specific influence of enlarged tonsils or adenoids on craniofacial growth [9]. To the best of our knowledge, no prior study in Vietnamese schoolchildren has simultaneously assessed AH and PTH separately, along with both palatal morphology and tongue strength, in the early mixed dentition.

The primary aim was to examine the individual and combined associations of AH and PTH with hard palate morphology and tongue strength in children with early mixed dentition. The exploratory objective was to examine whether an increasing adenotonsillar burden was associated with a composite phenotype of co-occurring orofacial alterations.

Materials and methods

The study was conducted in accordance with the Declaration of Helsinki, and its protocol was approved by the Institutional Review Board of Hue University of Medicine and Pharmacy (approval no. H2018/16). The observations were conducted and reported in accordance with STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. Prior informed consent was obtained in writing from all children and parents/caregivers.

Study design and participants

This is a cross-sectional study of 420 school-age children from 2 randomly selected primary schools in Hue city, Vietnam—one located in a rural area and the other in the central city—. In selected schools, all pupils aged 6–8 were invited to participate in the study, resulting in a response rate of 75%. Data collection took place from February to May in 2019. Inclusion criteria were children in the early mixed dentition with erupted permanent maxillary first molars. Exclusion criteria included loss of any maxillary reference teeth used for hard palate measurements (i.e., primary or permanent central incisors, primary canines and permanent first molars) due to caries or physiological exfoliation, a history of orthodontic treatment, adenotonsillectomy, and oral motor disabilities (dysphagia, facial or lingual nerve paralysis).

Epi-Info 7.0 software was used to calculate the minimum sample size for the study. The calculation was based primarily on estimating the prevalence of adenoid hypertrophy with adequate precision in the target population. Assuming an expected AH prevalence of 34% [1], a 95% confidence level, and an absolute precision of ± 5%, the minimum required sample was 345. This figure was inflated by approximately 10–12% to allow for potential non-response and incomplete records, yielding a target of 385. A total of 420 children were recruited, exceeding this target and ensuring precise estimation of the prevalence of adenotonsillar hypertrophy.

Calibration phase

A trained dentist performed clinical assessments of palatine tonsil grades, and a general dentist assessed tongue strength. Radiographic and arch model analyses were conducted by a single orthodontist who was unaware of the clinical findings. The theoretical and practical sessions of the training and calibration exercise were conducted in the orthodontics laboratory at Hue University of Medicine and Pharmacy.

The orthodontic calibration phase consisted of the orthodontist independently analyzing 30 dental casts and 30 cephalometric films randomly selected from the university’s orthodontic clinic patient records database (using available records from 2018). Intra-examiner variability was assessed by reanalyzing them after a 7-day interval. The assessment showed variability to be < 5% for all parameters. Intra-class correlation coefficients for radiographic and dental cast measurements exceeded 0.88, indicating a satisfactory level of intra-examiner reliability.

To prepare the principal investigator (a general dentist) for her task of assessing participants’ tonsil grade, an otolaryngologist was enlisted to assist in the calibration phase. Calibration commenced with a theoretical session to discuss the diagnostic criteria for tonsil grade, followed by practical training on 30 intraoral photographs of palatine tonsils. Assessments of the photographs by the dentist and the otolaryngologist showed high inter-examiner agreement (Kappa value: 0.96). The dentist reevaluated the photographs following a 7-day interval, yielding an intra-examiner reliability value of 0.93. During the training phase, the second investigator (also a general dentist) familiarized herself with the producer’s instructions for using the Iowa Oral Performance Instrument (IOPI, Medical LLC, Carnation, WA, USA) for tongue-strength testing, and practiced on members of the research team.

A pilot study was then conducted to evaluate all variables covered by the research. This study was carried out on a random sample of 30 children aged 6 to 8 from a single school. Participants of the pilot were excluded from the final sample. For clinical assessment, the principal investigator evaluated the palatine tonsil grade while the other dentist involved in the study assessed tongue strength. The reliability of tongue strength measurements was verified by repeating the assessment after a 7-day interval, yielding intra-class correlation coefficients exceeding 0.80, indicating strong reliability.

Data collection

Participants’ orofacial evaluation included assessments of tongue strength and, by dental cast analysis, of palatal morphology. Adenotonsillar status was evaluated clinically (using Brodsky’s grading system) [13] and radiographically (using Fujioka’s adenoid–nasopharynx ratio (ANR) from participants’ cephalometric images) [14].

Assessment of tongue strength

Maximal tongue strength was evaluated using the Iowa Oral Performance Instrument (IOPI, Medical LLC, Carnation, WA, USA) following Potter and Short’s (2009) validated procedure (Fig. 1) [15]. Participants applied maximum pressure for 3 s against a bulb positioned at midline of the tongue, with three trials conducted and the highest measurement recorded (in kilopascals (kPa)). The trials were separated by 30-second intervals in all sample groups. Tongue strength was recorded as the highest force peak across the three tests. Tongue strength was categorized as low (bottom 33%, < 39 kPa), medium (39–47 kPa), or high (top 33%, ≥ 48 kPa).

Fig. 1.

Assessment of Tongue Strength and Hard Palate Morphology. Legend: A Bulb position used for assessment and Iowa Oral Performance Instrument. B Arch model measurements obtained from digital dental-cast analysis (3Shape OrthoAnalyser): (a) intercanine width, (b) intermolar width, (c) anterior palatal length, (d) total palatal length, (e) anterior angle, (f) palatal height at canine area, (g) palatal height at molar area. Alt-text: Composite figure with two panels. Panel A shows the Iowa Oral Performance Instrument bulb positioned on the tongue for strength measurement. Panel B shows a digital dental cast with seven labeled linear and angular measurements of the maxillary arch and palate

Assessment of hard palate morphology

A digital model of the participant’s maxillary arch and palate (below, “arch model”) was created by scanning the corresponding plaster casts, and used to record linear and angular measurements in 3Shape OrthoAnalyser software (3Shape A/S, Copenhagen). All measurements were performed by one orthodontist, blinded to participants’ clinical characteristics. Intra-examiner variability was assessed by reanalyzing 30 randomly selected dental casts after 1 month, with variabilities < 5% for all arch parameters.

Hard palate morphology was assessed by measuring three-dimensional traits, including intermolar and intercanine width, palatal length, palatal height at the canine and molar area, as well as maxillary anterior angle (Fig. 1). Intermolar width was measured at the mesio-buccal cusp tips of first molars, and intercanine width at canine cusp tips. Palatal length measurements were taken from the contact point of upper central incisors to the line connecting the tips of the canines (anterior palatal length), and to the distal aspect of the upper first molars (total palatal length). Palatal height was measured vertically from mid-palatal suture to the horizontal plane intersecting mesio-palatal cusp tips of first molars (height at the molar area) and canine cusp tips (height at the canine area). Maxillary anterior angle (formed by lines drawn respectively from the tip of the right and the left canines to the contact point of the maxillary central incisors) was also measured. The measurements were used to calculate the following ratios:

1. intercanine width to anterior palatal length (canine W/L ratio);

2. intermolar width to total palatal length (molar W/L ratio);

3. palatal height at molar area to intermolar width (molar H/W ratio);

4. palatal height at canine area to intercanine width (canine H/W ratio).

Ratio 3 was multiplied by 100 to provide an index of palatine height, which was categorized into three height profiles based on the formula proposed by Maria et al.: low (≤ 27.9%), medium (28.0%–39.9%), and high (> 40.0%) [16].

An index of participants’ arch form (canine H/W ratio divided by molar H/W ratio) was computed using the formula proposed by Budiman, and arch forms were classified into three profiles according to his analysis: square arch (< 45.30%), oval arch (45.30%–53.37%), and tapered arch (> 53.37%) [17].

The palatal vault module was calculated as the half-sum of palatal height and the width of the alveolar part of the palate, ½(Pheight + Pwidth), providing a single index of overall vault size, with a smaller value indicating a narrower, more constricted vault [18]. 

Palatine tonsil size was assessed by visual inspection and graded according to Brodsky’s scale [13]. The Brodsky system categorizes tonsils into grades 0 to 4 based on the extent to which they obstruct the oropharyngeal airway: grade 0 indicates that tonsils are entirely within the tonsillar fossa, while grades 1–4 denote tonsils that occupy, respectively, < 25%, 26–50%, 51–75%, and > 75% of the airway (Fig. 2). To reduce the impact of possible asymmetry, tonsil size was assessed as the sum of the two sides of each tonsil, with grade 3 or 4 classed as PTH [13].

Fig. 2.

Assessment of Adenotonsillar Status. Legend: A Palatine tonsil size graded according to Brodsky's scale [13]: grade 0, tonsils within the tonsillar fossa; grade 1, < 25% airway obstruction; grade 2, 26–50%; grade 3, 51–75%; grade 4, > 75%. Grades 3 and 4 were classified as palatine tonsil hypertrophy (PTH). B Lateral cephalometric image illustrating adenoid size assessed using Fujioka's adenoid–nasopharynx ratio (ANR) [14]: A = distance from the outermost adenoid point to the spheno-basiocciput; N = distance from the spheno-basiocciput to the posterior nasal spine. ANR > 0.6 was classified as adenoid hypertrophy (AH). Alt-text: Composite figure with two panels. Panel A shows intraoral photographs illustrating Brodsky tonsil grades 0 through 4 with increasing oropharyngeal obstruction. Panel B shows a lateral cephalometric radiograph with lines A and N marked for calculation of the adenoid–nasopharynx ratio.

Adenoid size was assessed using lateral cephalometric radiographs obtained with the same X-ray machine (Soredex, Tuusula, Finland) according to a standard protocol (73 kV, 10 mA) by the same operator. All radiographs were taken in centric occlusion and natural head position without swallowing and traced digitally by an orthodontist using Dolphin Imaging^® 11.0 software. ANR was calculated following Fujioka et al.‘s method [14], dividing A (distance between the outermost adenoid point and spheno-basiocciput) by N (distance between spheno-basiocciput and posterior hard palate). AH was defined as ANR > 0.6 [19].

To assess the impact on orofacial structures of adenotonsillar obstruction patterns, participants were categorized into four groups based on obstruction type: AH only (ANR > 0.6; tonsil grade 0, 1, or 2), ATH (ANR > 0.6; tonsil grade 3 or 4), PTH only (ANR ≤ 0.6; tonsil grade 3 or 4), and normal (ANR ≤ 0.6; tonsil grade 0, 1, or 2).

Lip seal at rest was assessed by direct visual observation and recorded as competent (lips closed at rest) [20] or incompetent, with the degree of lip incompetence graded as mild, moderate, or severe [21]. Lip-seal status was entered in the models as an available proxy for resting oral posture and breathing tendency [20].

Outcomes

The primary outcome was the association between adenotonsillar status and orofacial characteristics, including hard palate morphology and posterior tongue strength. AH and PTH were first examined separately (present vs. absent, regardless of co-occurrence) to identify the orofacial measures associated with each obstruction type. A four-group classification (Normal, AH only, PTH only, combined ATH) was then used in multivariable models to test whether combined obstruction differed from either alone.

The exploratory outcome addressed whether the structural and functional domains of orofacial alteration — Linder-Aronson “adenoid facies” [5] reported in association with chronic upper airway obstruction [5] — co-occur more frequently with increasing adenotonsillar burden. Two composite definitions were analyzed: co-occurring alteration (both domains present simultaneously) and any-domain alteration (either domain present), providing complementary perspectives on specificity and sensitivity.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics version 22.0 (IBM Co., Armonk, NY, USA). Continuous variables were presented as mean ± standard deviation, and categorical variables as frequencies (percentages). No data were missing. Univariable comparisons by AH and PTH status used independent-samples t-tests or chi-square tests. To limit type I error, p-values were adjusted using the Benjamini–Hochberg false-discovery-rate procedure across all 22 comparisons.

For the primary objective, each continuous orofacial outcome was modeled by a separate multivariable linear regression in which adenotonsillar status (three dummy variables; Normal = reference), age, sex, BMI, lip-seal at rest, and school location were entered simultaneously. For the binary tapered-arch outcome, modified Poisson regression with robust variance was used to estimate adjusted prevalence ratios.

For the exploratory objective, both composite definitions (see Outcomes) were analyzed by modified Poisson regression with the same covariate set. A three-level adenotonsillar burden score (0 = neither; 1 = AH or PTH only; 2 = combined ATH) was entered as a continuous predictor to test for a graded trend. A two-sided p < 0.05 was considered significant for all analyses.

Results

Orofacial characteristics and adenotonsillar status

The study included 420 schoolchildren (48.1% female; mean age 7.61 ± 0.61 years; Table 1). Mean posterior tongue strength was 42.15 ± 13.55 kPa, and 35.0% of children were in the lowest tongue-strength tertile. A tapered arch was present in 31.0% and a high-arch palate in 20.0%. Enlargement of at least one pharyngeal lymphoid tissue was observed in 65.2% of participants (AH 32.6%, PTH 46.0%). The distribution across groups was: Normal 34.8% (n = 146), AH only 19.3% (n = 81), PTH only 32.6% (n = 137), and combined ATH 13.3% (n = 56).

Table 1.

Characteristics of the study population (N = 420)

Characteristic	N (%) or Mean (SD)
Demographics
Age (years)	7.61 (0.61)
Female sex, n (%)	202 (48.1)
School location (urban), n (%)	194 (46.2)
Anthropometrics
BMI (kg/m²)	15.89 (3.29)
BMI category, n (%)
Underweight	103 (24.5)
Normal	219 (52.1)
Overweight + Obesity	98 (23.4)
Adenotonsillar status
Adenoid hypertrophy (AH; ANR > 0.6)	137 (32.6)
Palatine tonsil hypertrophy (PTH; Brodsky 3/4)	193 (45.9)
Adenotonsillar group, n (%)
Normal	146 (34.8)
AH only	81 (19.3)
PTH only	137 (32.6)
Combined ATH	56 (13.3)
Maxillary arch morphology
Anterior angle (°)	122.87 (8.21)
Total palatal length (mm)	29.04 (2.08)
Canine W: AL ratio	4.57 (0.94)
Molar W: AL ratio	1.39 (0.13)
Palatal vault module	27.36 (1.69)
Tapered arch, n (%)	130 (31.0)
High-arch palate, n (%)	84 (20.0)
Tongue function
Posterior tongue strength (kPa)	41.89 (12.96)
Low tongue strength tertile, n (%)	147 (35.0)
Lip-seal & breathing
Nasal breathing (observed), n (%)	394 (95.4)
Complete lip-seal at rest, n (%)	237 (57.4)

SD Standard deviation, AH Adenoid hypertrophy, PTH Palatine tonsil hypertrophy, ATH Combined adenotonsillar hypertrophy

Comparison between children with and without ATH

In univariable analyses (Tables 2 and 3), AH was associated with a higher palatal vault (OR 1.67; 95% CI 1.01–2.75) and a shorter total palatal length (OR 0.88; 0.79–0.98). In the case of PTH, PTH was associated with a smaller anterior maxillary angle (OR 0.97; 0.94–0.99), a longer anterior palatal length (OR 1.19; 1.02–1.39), a narrower canine width-to-length ratio (OR 0.78; 0.61–0.98), a tapered arch (OR 1.78; 1.12–2.84), and lower tongue strength (OR 1.65; 1.01–2.72). None of the univariable comparisons remained significant after Benjamini–Hochberg correction (smallest q = 0.16), and pairwise post-hoc comparisons across the four adenotonsillar groups likewise showed no consistent pattern (Supplementary Table S1).

Table 2.

Hard palate morphology and tongue function by AH and PTH status

Variable	AH Yes (n = 137) M (SD)	AH No (n = 283) M (SD)	Δ	p	BH q	PTH Yes (n = 193) M (SD)	PTH No (n = 227) M (SD)	Δ	p	BH q
Maxillary arch parameters
Anterior angle (°)	122.67 (8.89)	122.96 (7.88)	−0.29	0.735	0.851	121.72 (7.73)	123.84 (8.50)	−2.12	0.008	0.160
Anterior length (mm)	7.31 (1.47)	7.33 (1.32)	−0.02	0.869	0.910	7.48 (1.35)	7.19 (1.38)	+ 0.29	0.029	0.160
Posterior length (mm)	28.73 (2.31)	29.19 (1.94)	−0.46	0.037	0.160	29.10 (2.09)	28.99 (2.07)	+ 0.11	0.575	0.748
Canine W: AL ratio	4.58 (1.08)	4.57 (0.86)	+ 0.01	0.949	0.949	4.46 (0.94)	4.67 (0.93)	−0.21	0.028	0.160
Molar W: AL ratio	1.40 (0.15)	1.39 (0.12)	+ 0.01	0.155	0.337	1.38 (0.13)	1.40 (0.13)	−0.02	0.065	0.179
Palatal vault
Palatal vault module	27.29 (1.82)	27.39 (1.62)	−0.10	0.578	0.748	27.14 (1.68)	27.54 (1.67)	−0.40	0.017	0.160
Arch form
Tapered arch, n (%)	45 (32.8)	85 (30.0)	+ 2.8	0.560	0.748	69 (35.8)	61 (26.9)	+ 8.9	0.050	0.160
Tongue function
Tongue strength (kPa)	40.68 (14.01)	42.47 (12.40)	−1.79	0.184	0.337	41.21 (13.50)	42.47 (12.48)	−1.26	0.321	0.543

Values are mean (SD) unless otherwise indicated. Δ = mean difference (Yes minus No); for tapered arch, Δ is the percentage-point difference. Bold p < 0.05 (nominal). No comparison reached Benjamini–Hochberg FDR threshold (smallest q = 0.160)

AH Adenoid hypertrophy, BH q Benjamini–Hochberg adjusted q-value, PTH Palatine tonsil hypertrophy, W:AL width-to-arch-length

Table 3.

Logistic regression odds ratios for association between hard palate morphology, tongue function, and AH or PTH

Variable	Odds ratio for AH†		Odds ratio for PTH†
	OR	95% CI	OR	95% CI
Maxillary arch parameters
Anterior angle	1.00	0.97–1.02	0.97	0.94–0.99
Anterior palatal length	1.00	0.85–1.17	1.19	1.02–1.39
Total palatal length	0.88	0.79–0.98	1.05	0.95–1.16
Canine W/L ratio	1.00	0.80–1.26	0.78	0.61–0.98
Molar W/L ratio	1.40	0.88–2.23	0.20	0.04–0.95
Arch form
Square	Ref.		Ref.
Oval	0.95	0.57–1.60	1.57	0.97–2.56
Tapered	1.20	0.74–1.96	1.78	1.12–2.84
Palate height
Low or medium	Ref.		Ref.
High	1.67	1.01–2.75	0.99	0.61–1.60
Tongue strength
Low	1.07	0.64–1.78	1.65	1.01–2.70
Medium	0.86	0.51–1.45	1.13	0.69–1.86
High	Ref.		Ref.

Bold OR indicates 95% CI excludes 1.00. Each variable entered in a separate univariable logistic model

W/L Width-to-length ratio

† Adjusted for sex, age group, BMI group

In the adjusted multivariable models (Table 4), combined ATH was associated with a smaller anterior maxillary angle (β = −3.76°; 95% CI − 6.22 to − 1.29; p = 0.003) and a smaller palatal vault module (β = −0.54; 95% CI − 1.02 to − 0.05; p = 0.032). The canine width-to-length ratio showed a consistent direction but did not reach significance (β = −0.28; 95% CI − 0.57 to 0.01; p = 0.057). Both palatal measures decreased with increasing adenotonsillar burden (anterior angle β = −1.54 per level, p = 0.010; vault module β = −0.27 per level, p = 0.020). Among the covariates, age and urban school location were the strongest independent correlates of posterior tongue strength (β = +7.29 and + 6.64 kPa, respectively; both p < 0.001). A pre-specified school-stratified analysis revealed that the association of combined ATH with lower tongue strength was confined to the rural setting (rural β = −7.45 kPa, p = 0.006; urban β = +2.90, p = 0.279; interaction p = 0.002; Supplementary Table S2).

Table 4.

Adjusted multivariable associations between adenotonsillar status and individual continuous orofacial measures (N = 420)

Predictor	Anterior angle (°) β (95% CI); p	Vault module β (95% CI); p	Canine W: AL ratio β (95% CI); p	Tongue strength (kPa) β (95% CI); p
Adenotonsillar status (ref = Normal)
AH only	0.95 (− 1.23, 3.12); 0.391	−0.09 (− 0.53, 0.34); 0.667	0.07 (− 0.18, 0.32); 0.586	0.28 (− 3.00, 3.57); 0.865
PTH only	−1.43 (− 3.31, 0.46); 0.137	−0.40 (− 0.78, − 0.03); 0.035	−0.14 (− 0.36, 0.07); 0.195	−0.64 (− 3.48, 2.21); 0.660
Combined ATH	−3.76 (− 6.22, − 1.29); 0.003	−0.54 (− 1.02, − 0.05); 0.032	−0.28 (− 0.57, 0.01); 0.057	−2.80 (− 6.51, 0.92); 0.140
Per burden level (trend)	−1.54 (− 2.71, − 0.37); 0.010	−0.27 (− 0.50, − 0.04); 0.020	—	—
Covariates
Age, years	−0.10 (− 0.55, 0.34) ; 0.547	0.33 (0.07, 0.59); 0.014	0.05 (− 0.02, 0.12) ; 0.227	7.29 (5.28, 9.30); <0.001
Male sex	0.27 (− 0.97, 1.51) ; 0.073	0.32 (0.01, 0.63); 0.041	−0.09 (− 0.27, 0.09) ; 0.219	−1.32 (− 3.67, 1.04) ; 0.255
BMI	−0.27 (− 0.53, − 0.02); 0.036	0.17 (0.12, 0.22); <0.001	−0.04 (− 0.07, − 0.01); 0.020	0.29 (− 0.10, 0.68) ; 0.108
School: Urban	−2.51 (− 4.23, − 0.78); 0.005	−0.12 (− 0.45, 0.21); 0.999	−0.08 (− 0.27, 0.11) ; 0.343	6.64 (4.04, 9.25); <0.001
R² (adjusted)	0.067	0.130	0.028	0.146

Values are unstandardised β coefficients (95% confidence interval); exact p shown where p < 0.05. Adjusted multivariable linear regression (covariates: age, sex, BMI, lip-seal at rest, school location). Lip-seal estimates were non-significant and are omitted. Per-burden-level estimates are from separate models entering the 3-level adenotonsillar burden score (0/1/2) as a continuous predictor

W:AL width-to-arch-length ratio

Orofacial Alterations and Presence of ATH

Co-occurring alteration (both domains; 77 children, 18.3%): prevalence rose from 11.6% to 25.0% with increasing burden. Modified Poisson regression showed significant associations with PTH only (aPR 1.77; 95% CI 1.03–3.06; p = 0.040) and combined ATH (aPR 1.87; 95% CI 1.01–3.46; p = 0.048), with a graded trend (aPR 1.38 per level; p = 0.020) (Table 5; Fig. 3).

Table 5.

Co-occurring orofacial alteration by adenotonsillar status: modified poisson regression (N = 420)

Adenotonsillar status	n (%)	Co-occurrence, n (%)	aPR (95% CI)	p
Normal	146 (34.8)	17 (11.6)	1.00 (ref)	—
AH only	81 (19.3)	17 (21.0)	1.66 (0.90–3.04)	0.104
PTH only	137 (32.6)	29 (21.2)	1.77 (1.03–3.06)	0.040
Combined ATH	56 (13.3)	14 (25.0)	1.87 (1.01–3.46)	0.048
Per burden level (trend)	—	—	1.38 (1.05–1.82)	0.020

Co-occurring alteration = structural hard palate alteration (tapered arch or high-arch palate) and low posterior tongue strength (lowest tertile); 77/420 (18.3%). aPR adjusted prevalence ratio (modified Poisson, robust variance); adjusted for age, sex, BMI, lip-seal at rest, school location. Burden score: 0 = neither, 1 = AH or PTH only, 2 = combined ATH. See Supplementary Table S3 for the any-domain definition

Bold values indicate statistically significant associations (p < 0.05)

Fig. 3.

Distribution of orofacial alteration categories by adenotonsillar burden Legend: Children were classified into three categories based on the presence of structural hard palate alteration (tapered arch or high-arch palate) and/or low posterior tongue strength (lowest tertile): no alteration (neither domain affected), single-domain alteration (one domain only), or co-occurring alteration (both domains). Adenotonsillar burden: 0 sites = Normal (n = 146); 1 site = AH only or PTH only (n = 218); 2 sites = combined ATH (n = 56). The proportion with co-occurring alteration rose from 11.6% to 21.1% to 25.0% across burden levels. Alt-text: Stacked bar chart showing three orofacial alteration categories (no alteration, single-domain, co-occurring) across three adenotonsillar burden levels. The co-occurring category increases progressively from 11.6% at zero obstructive sites to 21.1% at one site and 25.0% at two sites

Any-domain alteration (either domain; 267 children, 63.6%): a consistent graded trend was observed (aPR 1.12 per burden level; 95% CI 1.01–1.24; p = 0.039), although individual group estimates did not reach significance (Supplementary Table S3). Across both outcome definitions, the prevalence of orofacial alterations was higher at greater adenotonsillar burden levels.

Discussion

This cross-sectional study of 420 prepubertal Vietnamese schoolchildren aged 6–8 years examined how adenotonsillar status relates to hard palate morphology and tongue strength, distinguishing AH, PTH and combined ATH. Combined ATH and greater adenotonsillar burden were associated with a smaller anterior maxillary angle and palatal vault module, consistent with transverse maxillary constriction. AH and PTH showed different univariable orofacial patterns, but these did not survive FDR correction. A secondary co-occurring phenotype was more frequent with greater burden, though this exploratory outcome requires confirmation.

Orofacial characteristics and adenotonsillar status

The study highlighted a notable prevalence of ATH among schoolchildren with mixed dentition: 65.2% of the sample exhibited enlargement in at least one pharyngeal lymphoid tissue. Although hospital-based studies report higher AH prevalence rates (ranging from 42% to 70% in children and adolescents) [22], our findings (AH prevalence of 32.62%) align with those of other paediatric population studies [1]. PTH was present in 45.9% of children, higher than the 11% reported in Turkish schoolchildren [2]. In terms of orofacial functions, our study found that approximately one-third of children from the general community exhibit various alterations, including tapered arch form (31.0%), high-arch palate (20.0%), and low tongue pressure against hard palate (35.0%). Although not the primary aim, these data contribute to the limited pediatric evidence on tongue strength in 6–8-year-old children. The mean posterior tongue-strength value was approximately 42 kPa, which is slightly lower than previously reported values in typically developing children of comparable age (46 kPa at age 6) [23]. School location was independently associated with posterior tongue strength, and the inverse association of combined ATH with tongue strength appeared confined to the rural setting (interaction p = 0.002; Supplementary Table S4). School location is a broad contextual marker potentially reflecting nutrition, environmental exposures, and access to care; and is best interpreted as a proxy.

The situation described above suggests a neglect of oral health (the latter being defined by WHO’ as “the state of the mouth, teeth, and orofacial structures that enable individuals to perform essential functions”) [24]. Our study population showed a high occurrence of adenoid and palatine tonsil hypertrophy with upper airway obstruction as well as orofacial alterations in hard palate morphology and tongue strength, which indicates an underdiagnosis of oral health issues in 6–8-year-old schoolchildren. ATH is part of the physiology of craniofacial growth and development in healthy children, peaking between ages 3–8 and regressing by 8–10, with complete atrophy by 12–14 years [1]. Palatal growth begins around 6–9 years, along with facial growth [1], and the tongue muscle continues to develop in volume and elasticity until puberty [23]. Since upper airway obstruction may recede and orofacial function improve with the onset of maturity during mixed dentition, our study was targeted to pre-puberty schoolchildren with an age range of only two years and we adjusted our analysis for participants’ age to minimize the impact of maturation. At this stage, the bone remodeling process caused by oral respiration has still not reached its maximum [25]. Understanding the etiological factors and co-morbidities or orofacial problems in this age group is important for identifying and seizing early treatment opportunities, allowing growth during puberty to compensate for alterations [25], and enhancing the long-term stability of eventual interventions [8].

Comparison of children with and without ATH

In unadjusted analyses, children with AH nominally showed a shorter total palatal length and a higher-arch palate, although these findings did not survive FDR correction. They are directionally consistent with a Nigerian study describing a similar altered maxillary growth pattern in children with AH (a smaller maxilla with shorter arch length and a steep palatal vault) [26]. These observations partly resemble the “adenoid face” phenotype described by Linder-Aronson (long, narrow, with a high-arch palate and retrognathic mandible) [5]; in our data, AH-associated nasopharyngeal obstruction was related to altered craniofacial morphology (a smaller maxilla and high-arch palate). However, our data did not establish a correlation between AH and transverse constriction of the maxillary, which suggests that AH alone may not predict altered arch form accurately.

Nominal shifts in transverse arch parameters were observed in PTH cases. Nominally, children with PTH tend to have a narrower, tapered arch with a smaller anterior angle and longer anterior length, although these findings did not survive FDR correction. The tapered arch typically narrows towards anterior teeth, resembling the narrow anterior segment of V-shaped arches suggested by Thompson [27]. Other authors have also found that oropharynx obstruction associated with PTH inhibits transverse growth of the maxilla [4, 8].

While AH in our study did not have a statistically relevant relationship to low tongue strength, the nominal relationship was present for PTH, which appears to be associated with a weakening of tongue muscles. Reduced tongue strength may be clinically relevant because tongue pressure contributes to stomatognathic function and has been discussed in relation to speech, mouth breathing, and pediatric sleep-disordered breathing [19, 23]. Tongue-strength measurement using the IOPI may therefore be useful as part of broader orofacial functional screening, but interventional implications require longitudinal and clinical confirmation [15, 22, 29].

Overall, AH and PTH showed different exploratory patterns of association with hard palate morphology and tongue function. PTH was more closely related to transverse hard palate dimensions and lower tongue strength, and AH was more closely related to vertical and sagittal palatal dimensions. These patterns were exploratory and did not survive false-discovery-rate correction; they are therefore reported as hypothesis-generating rather than established distinct phenotypes.

Orofacial alterations and presence of ATH

The alterations most often associated with Waldeyer’s hypertrophy included narrow tapered arch, high-arch palate, and reduced tongue strength or hypotonia, typical features of an “adenoid face” [5]. Recent studies have raised the question of whether more severe upper-airway obstruction may intensify orofacial alterations, particularly in children with concurrent adenoid and palatine tonsil hypertrophy [8, 11]. In our data, this phenotype became more prevalent with increasing adenotonsillar burden, and modified Poisson regression showed a significant association with combined ATH. These associations were modest in magnitude and are best interpreted as describing the frequency with which the two domains coincide, rather than as evidence of a causal or amplifying relationship.

Kang et al. proposed, in a pediatric obstructive-sleep-apnea context, that combined adenotonsillar enlargement may contribute more strongly to airway obstruction than either component alone [12]. Our finding aligns with the consensus in the literature regarding ATH as the predominant factor in upper airway obstruction and obstructive sleep apnea among children [19, 30]. Nevertheless, the effect sizes were modest and the confidence intervals were relatively wide; therefore, the result should be viewed as hypothesis-generating and useful for planning longitudinal or multidisciplinary studies rather than as evidence for clinical prediction.

The individual continuous measures (anterior maxillary angle, palatal vault module) carried the more robust evidence for the structural associations. The palatal vault module — the half-sum of palatal height and width [18] — is normally stable across gnathic types (~ 27 mm). Because width exceeds height by approximately 2.4-fold in this age group, the module is width-dominated; the smaller value observed in children with combined ATH therefore primarily reflects transverse constriction rather than a shallower vault. Together with the smaller anterior maxillary angle and the univariable association of AH with a higher palatal height index, these findings describe a high but narrow vault consistent with the “adenoid facies”, although this interpretation remains exploratory. The co-occurring phenotype provides a complementary structural–functional descriptor for future confirmatory study. Pediatric dentists can play an important role in early detection and timely referral, as demonstrated by the feasibility of orofacial dysfunction screening in children with sleep-disordered breathing symptoms [31]. While the causal relationship remains uncertain, a comprehensive assessment should be conducted alongside the management of adenotonsillar and associated alterations. Attention should not be focused solely on ATH-associated airway obstruction; it is crucial to consider preventing and treating maxillary deformity and muscular impairment co-morbidities during the initial stage of dental arch development. A multidisciplinary team (pediatrician, otolaryngologist, pediatric dentist/orthodontist, and myofunctional therapist) could coordinate timely ENT referral, early orthodontic evaluation, and oral motor function assessment.

Limitations

Several limitations should be acknowledged. Adenoid size was assessed using the adenoid–nasopharynx ratio on lateral cephalometric radiographs. Although ANR is widely used in orthodontic and pediatric airway studies, it is a two-dimensional proxy for a three-dimensional lymphoid structure and may be less accurate than nasoendoscopy or CBCT. Palatine tonsil size was assessed clinically using Brodsky grading; examiner calibration was performed, but some subjectivity remains. The cross-sectional design precludes causal inference or determination of temporal sequence among adenotonsillar status, palatal morphology, tongue strength, and breathing pattern. The study at hand did not assess the lingual tonsil—this suggests a promising direction for further clinical research. Our relatively large sample and our random sampling of children in the mixed dentition age bracket from the general population rather than from clinical settings or specific-pathology groups considerably enhances the representativeness of our findings for schoolchildren.

Residual confounding is also possible. Lip-seal at rest and school location were included as available proxies for breathing tendency and contextual factors, but detailed data on habitual mouth breathing, allergic rhinitis, parental education, and objectively measured sleep-disordered breathing were not available. The low tongue-strength category was sample-derived because validated age- and population-specific clinical cut-offs are not available for Vietnamese children aged 6–8 years. Finally, the univariable comparisons did not survive false-discovery-rate correction and should be regarded as hypothesis-generating. The co-occurring phenotype was exploratory and should not be interpreted as a validated clinical diagnosis. Because the study was conducted in one region of Vietnam, generalisability to other populations should be interpreted cautiously.

Conclusion

The study highlights a high prevalence of adenotonsillar hypertrophy in schoolchildren. AH and PTH showed distinct patterns of hard palate and tongue-function alteration, although these exploratory distinctions did not survive false-discovery-rate correction. The most consistent evidence supported anterior and transverse maxillary constriction in children with combined ATH and greater adenotonsillar burden. Co-occurring hard palate alteration and low posterior tongue strength were more frequent with greater adenotonsillar burden. These findings support multidisciplinary orofacial screening during the mixed-dentition period, while longitudinal studies are needed to clarify temporality and clinical implications.

Authors' contributions

D.A.H., T.M.N, V.T.N.N. and T.J. conceived the ideas and designed; D.A.H., D.N.T.N. and T.A.H. collected the data; D.A.H., T.A.H. and V.T.N.N. analyzed the data; D.A.H., T.M.N. and T.J. led the writing; and D.A.H., T.M.N., V.T.N.N., D.N.T.N., T.A.H. and T.J. critically revised the manuscript. All authors gave their final approval and agree to be accountable for all aspects of the work.

Funding

No funding was received to assist with the preparation of this manuscript.

Data availability

The datasets are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the Bioethics Committee of the Hue University of Medicine and Pharmacy, Vietnam (No. H2018/16).

Written informed consent was obtained from all children and parents/caregivers.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.