We conducted a retrospective review of the electronic medical records of children aged 6–12 years who underwent elective adenoidectomy and/or tonsillectomy at the Otorhinolaryngology Department of the First Affiliated Hospital of Zhengzhou University. We extracted data from the first page of medical records spanning January 2018 to August 2022, identifying cases using the International Classification of Diseases (ICD-10), along with the Current Procedural Terminology and Healthcare Common Procedure Coding System codes. A total of 8,583 patients were admitted during this period, of whom 3,026 (1,938 boys and 1,088 girls) met the inclusion and exclusion criteria (Figure 1). The study was conducted in accordance with the Declaration of Helsinki, approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Approval No: 2022-KY-1335). As this research was based on a retrospective review of existing medical records and involved no direct patient intervention, and as all data were fully anonymized before analysis, the requirement for individual informed consent was waived by the Ethics Committee.

The inclusion criteria for the study were as follows: (1) age between 6 and 12 years; (2) first-time hospitalization at the First Affiliated Hospital of Zhengzhou University; (3) availability of routine preoperative test results, including complete blood count, urinalysis, electrolyte levels, liver enzymes [alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT)], and kidney function markers [serum creatinine (Cr) and blood urea nitrogen (BUN)], all within normal ranges adjusted for sex and age; (4) thyroid function markers (free triiodothyronine, free thyroxine, and thyroid-stimulating hormone) within normal ranges for sex and age; and (5) adenoidectomy conducted due to adenoid and tonsillar hypertrophy.

The exclusion criteria for the study were as follows: (1) incomplete electronic records, especially missing laboratory data; (2) a history of arthritis or joint pain, including gout; (3) presence of endocrine disorders or syndromes impacting metabolic function, such as conditions of the endocrine pancreas, pituitary, adrenal glands, thyroid, or parathyroid gland; (4) a history of regular medical follow-ups or routine medication use; (5) adenoidectomy performed due to recurrent ear infections or otitis; and (6) a history of prior adenoidectomy, revision surgery, or any other major surgical procedure.

Demographic information, including age, sex, weight, and height, was collected from all participants. The body mass index (BMI) z-score was calculated based on the World Health Organization (WHO) growth standards for children aged 2–5 years and the WHO growth reference for those aged 5–19 years using the Age-based Pediatric Growth Reference Charts website (10).

All blood samples were obtained after overnight fasting as part of routine preoperative laboratory testing. Biochemical parameters included serum uric acid (SUA), fasting plasma glucose (FPG), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP), BUN, and creatinine (Cr). These measurements were performed using a standardized automatic biochemistry analyzer (Roche, Germany). Hematological parameters were obtained from a complete blood count (CBC), including white blood cell (WBC), neutrophil, lymphocyte, and monocyte counts. These hematological indices were further used to calculate the derived inflammatory markers. CBC testing was performed using an automated hematology analyzer (Mindray, China) following standardized operational protocols. Internal quality control procedures were applied throughout the testing process, and the coefficient of variation remained below 5%. The following CBC-derived inflammatory indicators were calculated: NLR, dNLR, MLR, NMLR, SIRI, and SII, using the following formulas: NLR = neutrophil counts/lymphocyte counts, dNLR = neutrophil counts/(white blood cell counts − lymphocyte counts), MLR = monocyte counts/lymphocyte counts, NMLR = (monocyte counts + neutrophil counts)/lymphocyte counts, SIRI = neutrophil counts × monocyte counts/lymphocyte counts, and SII = platelet counts × neutrophil counts/lymphocyte counts. The TG/HDL-C ratio was calculated as TG (mmol/L) divided by HDL-C (mmol/L) (11).

We categorized all participants into quartiles based on their serum uric acid (SUA) levels: Q1 (SUA ≤ 3.66 mg/dL), Q2 (3.66 mg/dL < SUA ≤ 4.28 mg/dL), Q3 (4.28 mg/dL < SUA ≤ 5.02 mg/dL), and Q4 (SUA > 5.02 mg/dL).

Participants were defined to have dyslipidemia if they met any of the following conditions: TC ≥ 5.20 mmol/L; TG ≥ 1.10 mmol/L for children younger than 10 years or ≥1.50 mmol/L for those aged 10 years or older; HDL-C < 1.00 mmol/L; LDL-C ≥ 3.40 mmol/L; or non-HDL-C ≥ 3.70 mmol/L (12).

Continuous variables with a normal distribution are expressed as mean ± standard deviation (mean ± SD), whereas non-normally distributed continuous variables are presented as median (P25, P75). Categorical variables are summarized as percentages. For multiple comparisons of normally distributed variables, pairwise between-group analyses were conducted using the Bonferroni correction when variances were homogeneous and the Dunnett T3 method when variances were unequal. For multiple comparisons of non-normally distributed variables, the Kruskal–Wallis H test was employed, followed by post-hoc pairwise comparisons using the Dunn–Bonferroni correction. Spearman’s test was used to calculate the correlation coefficient (rho) between the TG/HDL-C and clinical parameters. Correlation values of p value of <0.05 were considered statistically significant. Restricted cubic spline (RCS) regression models were used to examine the nonlinear relationship between SUA and the TG/HDL-C ratio. The RCS model incorporated 4 knots placed at the 5th, 35th, 65th, and 95th percentiles of the variable distribution. All RCS analyses were adjusted for potential confounders, including age, BMI z-score, sex, ALT, AST, GGT, and ALP levels. To further quantify the inflection point suggested by the RCS models, we fitted two piecewise linear regression models using data-driven grid search. For each candidate SUA value (increment of 0.1 mg/dL), we constructed a two-segment linear model and selected the inflection point that minimized the Akaike Information Criterion (AIC). Non-parametric bootstrap resampling (500 iterations) was used to estimate the sampling distribution of the inflection point and to derive the 95% confidence intervals. All statistical tests were two-sided, and a p value of <0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 21.0 for Windows (SPSS Inc., Chicago, IL) and R version 4.4.1.¹

The baseline characteristics of the study participants are summarized in Table 1. A total of 3,026 participants were included, with approximately 64.0% being male. The mean age of the cohort was 8.4 ± 1.9 years, and the overall mean SUA level was 4.41 ± 1.06 mg/dL. Mean SUA levels for each quartile were as follows: Quartile 1, 3.20 ± 0.36 mg/dL; Quartile 2, 3.98 ± 0.17 mg/dL; Quartile 3, 4.62 ± 0.21 mg/dL; and Quartile 4, 5.83 ± 0.77 mg/dL. Participants’ age increased progressively across SUA quartiles (p < 0.001). There was no significant difference in the male-to-female ratio between the quartile groups (p = 0.060). The mean TG/HDL-C ratio was 1.64 ± 1.38, with values of 1.37 ± 1.12, 1.48 ± 1.15, 1.65 ± 1.38, and 2.07 ± 1.6 across the quartiles (p < 0.001). As the SUA quartile increased, significant upward trends were observed for several metabolic markers, including BMI z-score, Cr, BUN, TC, TG, LDL-C, and ALT (p < 0.050 for all). Conversely, HDL-C levels showed a significant decrease with increasing SUA levels (p < 0.001). Notably, FPG levels showed no significant variation across quartiles (p = 0.096).

As shown in Table 1, increasing SUA quartiles from Q1 to Q4 were accompanied by a clear upward trend in multiple hematological parameters. Participants in the higher SUA quartiles showed progressively higher WBC, neutrophil, monocyte, and platelet counts. Consistently, all CBC-derived inflammatory indices, including NLR, dNLR, MLR, NMLR, SIRI, and SII, demonstrated a stepwise increase across SUA quartiles, suggesting a gradual enhancement of the systemic inflammatory status associated with elevated SUA levels.

Table 2 presents the correlations between clinical parameters and the TG/HDL-C ratio. In the overall population, TG/HDL-C showed significant positive correlations with age (r = 0.201, p<0.001), BMI z-score (r = 0.168, p < 0.001), SUA (r = 0.232, p < 0.001), LDL-C (r = 0.152, p < 0.001), and liver enzyme levels, including ALT (r = 0.171, p < 0.001), ALP (r = 0.138,) and GGT (r = 0.277, p < 0.001). Conversely, AST exhibited a modest negative correlation with the TG/HDL-C ratio (r = −0.176, p < 0.001).

In addition, hematological parameters (including WBC, neutrophil count, monocyte count, and platelet count), as well as CBC-derived inflammatory indices, such as NLR, dNLR, MLR, SIRI, and SII, showed a significantly positive correlation with the TG/HDL-C ratio (all p < 0.050).

A restricted cubic spline (RCS) regression model was applied to conduct a piecewise analysis of the relationship between SUA and the TG/HDL-C ratio (Figure 2). The analysis revealed a nonlinear positive association between SUA and the TG/HDL-C ratio (p < 0.001, p for nonlinear = 0.008). When SUA < 4.40 mg/dL, a modest positive correlation with the TG/HDL-C ratio was observed (β: 0.14, 95% CI: 0.04 ~ 0.25, p = 0.006). When SUA ≥ 4.40 mg/dL, the correlation became significantly stronger (β: 0.40, 95% CI: 0.32 ~ 0.47, p < 0.001). When stratified by the SUA inflection point (4.4 mg/dL), the proportion of participants with a BMI z-score >2 and the prevalence of dyslipidemia were markedly higher in the higher SUA group than in the lower SUA group (Table 3). Collectively, these findings imply that elevated SUA beyond the 4.40 mg/dL threshold may serve as an early marker for increased susceptibility to overweight and dyslipidemia in children with adenoid or tonsillar hypertrophy.

To further evaluate the robustness of the SUA inflection point, we applied data-driven two-piecewise linear regression with non-parametric bootstrap resampling (500 iterations). The bootstrap analysis yielded a mean inflection point of 4.61 mg/dL, with a 95% CI of 3.28–5.79 mg/dL, supporting the stability of the nonlinear transition region identified in the RCS model. Importantly, the bootstrap-derived confidence intervals overlapped with the inflection regions identified by the RCS models, supporting the robustness and consistency of the observed nonlinear relationship between SUA and the TG/HDL-C ratio.

The RCS linear regression model was employed to conduct a piecewise regression of the association between SUA and the TG/HDL-C ratio across different sexes. The relationship between SUA and the TG/HDL-C ratio in boys and girls aligns with the pattern observed in the general population (boys: p = 0.002, p for nonlinear = 0.018; girls: p<0.001, p for nonlinear = 0.029). For boys, when SUA < 4.60 mg/dL, the TG/HDL-C ratio exhibited a modest positive correlation (β: 0.15, 95% CI: 0.03 ~ 0.27, p = 0.012). When SUA ≥ 4.60 mg/dL, the correlation became significantly stronger (β: 0.39, 95% CI: 0.29 ~ 0.49, p < 0.001) (Figure 3). For girls, when SUA < 3.75 mg/dL, there was no statistically significant correlation with the TG/HDL-C ratio (β: −0.07, 95% CI: −0.33 ~ 0.25, p = 0.18). In contrast, when SUA ≥ 3.75 mg/dL, a statistically significant positive correlation with the TG/HDL-C ratio was identified (β: 0.44, 95% CI: 0.34 ~ 0.53, p < 0.001) (Figure 4).

To further assess the robustness of the sex-specific inflection points observed in the RCS analyses, we conducted bootstrap-based two-piecewise linear regression analyses (500 iterations) for boys and girls separately. Among boys, the bootstrap procedure yielded a mean inflection point of 4.61 mg/dL (95% CI: 3.28–5.79 mg/dL). Among girls, the corresponding mean inflection point was 3.94 mg/dL (95% CI: 3.27–5.35 mg/dL).