Adenoid-Nasopharyngeal Ratio and craniofacial growth in Children: A two-center longitudinal study with threshold and causal analyses.

purposeAdenoid hypertrophy is a common cause of pediatric upper-airway obstruction and is suspected to contribute to "adenoid facies", but longitudinal thresholds and causal dose-response relations between the adenoid-nasopharyngeal (A/N) ratio and craniofacial growth indices remain unclear.

methodsWe performed a two-center longitudinal cohort study of children aged 3-10 years who underwent clinically indicated lateral nasopharyngeal or cephalometric radiography between July 2021 and July 2024 at tertiary pediatric hospitals in Shanghai and Chongqing, China. Children with at least two eligible radiographic visits were included. The exposure was the distance-based A/N ratio at the earlier visit. Outcomes were annualized changes in standard cephalometric angles and mesh-based vertex angles, with emphasis on cranial-base and posterior maxillary/nasopharyngeal angles. Thresholds were screened using single-break segmented regression, and causal dose-response functions were estimated using generalized propensity score models stratified by sex and age.

resultsA total of 4,271 children contributed 4,907 temporally ordered within-child radiograph visit pairs (each pair links an earlier and a later visit; children with ≥ 3 visits contribute multiple pairs). The median baseline A/N ratio was 0.601 (interquartile range, 0.549-0.652). Across 189 angles, segmented regression identified a reproducible breakpoint band around A/N ≈ 0.62-0.69. Generalized propensity score analyses showed positive, approximately linear dose-response functions for multiple cranial-base and posterior maxillary/nasopharyngeal angles, with global slopes typically 10-25°/year per 1.0 increase in A/N and 80th-20th percentile contrasts of about 1-4°/year. Baseline two-group comparisons showed synchronous reductions of about 3° in SNA (sella-nasion-A angle) and SNB (sella-nasion-B angle) in children with adenoid hypertrophy, with ANB (A-nasion-B angle) largely unchanged.

conclusionHigher radiographic A/N ratios were associated with threshold-sensitive, age- and sex-dependent craniofacial remodeling. As a 2D radiographic surrogate, A/N does not replace endoscopic or functional assessment. Combining A/N with selected cephalometric angles may help identify children who warrant closer follow-up and endoscopic evaluation during sensitive growth windows.