Abstract / Summary
Abstract Background We assessed the association between the non-high-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio (NHHR) and prevalent hyperuricemia (HUA) in adults hospitalized with fractures. We also examined overlap with other lipid measures and exploratory incremental discrimination. Methods This retrospective cross-sectional study included 2699 adults aged ≥ 50 years with observed serum uric acid, total cholesterol, and high-density lipoprotein cholesterol (HDL-C). Each patient contributed one index hospitalization, defined as the first eligible hospitalization during the study period. Fasting biochemical measurements were obtained within the first 24 hours after admission; later inpatient values were excluded. HUA was defined as serum uric acid ≥ 416 μmol/L in men or ≥357 μmol/L in women. We estimated prevalence odds ratios (ORs) and modified-Poisson prevalence ratios (PRs) as complementary measures of association across 20 multiple-imputation (MI) datasets. Models used clinically informed covariates and flexible estimated glomerular filtration rate (eGFR) adjustment. Complete-case, selection-weighted, and alternative-threshold analyses assessed sensitivity. Results HUA was present in 418 patients (15.5%). Per one standard deviation (SD) higher NHHR, the MI-pooled OR was 1.188 (95% confidence interval [CI] 1.061–1.330; p = 0.003). The corresponding PR was 1.129 (95% CI 1.045–1.220; p = 0.002). Selection-weighted estimates were similar (OR 1.191; PR 1.131), although lipid missingness and limited positivity constrain interpretation. The complete-case OR was 1.130 (95% CI 0.999–1.277). Nonlinear evidence was unstable after trimming extreme NHHR values. In a joint model including triglycerides and low-density lipoprotein cholesterol, the NHHR OR was 1.046 (95% CI 0.876–1.248). Adding NHHR changed cross-validated area under the receiver operating characteristic curve (AUC) from 0.777 to 0.780 (ΔAUC 0.0026; conditional 95% interval −0.0016–0.0070). Conclusions NHHR showed a modest association with prevalent HUA in this selected fracture-hospitalized population. Interpretation is tempered by missing-data assumptions, overlapping lipid information, and limited incremental discrimination. Fasting first-24-hour sampling cannot fully distinguish chronic metabolic status from acute hospitalization physiology; no validated prediction role or clinical NHHR cutoff is established. Clinical trial number Not applicable.