Stress hyperglycemia ratio and albuminuria in Shandong and Jiangsu, China: a two-center cross-sectional study with inter-hospital external evaluation

The stress hyperglycemia ratio (SHR) expresses contemporaneous glycemia relative to the chronic glycemic background. In clinically stable health-examination populations, SHR should be interpreted as relative fasting glycemic elevation rather than evidence of acute stress hyperglycemia. We examined whether SHR was reproducibly associated with same-visit albuminuria across two independent hospital health-examination cohorts. This two-center cross-sectional study included 2,378 adults who underwent a single-visit health examination and questionnaire assessment at the Second Affiliated Hospital of Shandong First Medical University (Shandong cohort; n = 1,311) and Taizhou People’s Hospital Affiliated to Nanjing Medical University (Taizhou cohort; n = 1,067) during 2024–2025. SHR was calculated as fasting plasma glucose divided by hemoglobin A1c-derived estimated average glucose. The primary outcome was albuminuria, defined as urinary albumin-to-creatinine ratio (UACR) ≥ 30 mg/g. The primary analysis examined the SHR-albuminuria association using multivariable regression and restricted cubic splines; sensitivity and subgroup analyses assessed robustness. Inter-hospital evaluation of coefficient-locked models was retained as a secondary same-visit risk-enrichment analysis rather than prospective prediction. Albuminuria on the study visit was present in 323 participants (13.58%): 179 (13.65%) in the Shandong cohort and 144 (13.50%) in the Taizhou cohort ( P = 0.911). Each 1-SD increase in SHR was associated with higher odds of albuminuria after full adjustment in the pooled sample (OR, 1.65; 95% CI, 1.46–1.87; P < 0.001), Shandong cohort (OR, 1.67; 95% CI, 1.39–2.01; P < 0.001), and Taizhou cohort (OR, 1.69; 95% CI, 1.42–2.01; P < 0.001), without center interaction (P for interaction = 0.873). Albuminuria prevalence increased from 6.89% in SHR quartile 1 to 21.51% in quartile 4; the fully adjusted OR for quartile 4 versus quartile 1 was 3.92 (95% CI, 2.63–5.83; P < 0.001). Dose-response associations were approximately linear in the pooled and center-specific analyses. In the secondary inter-hospital same-visit classification analysis, adding SHR increased the Taizhou-cohort AUC from 0.75 to 0.78 (absolute delta AUC, 0.03; P = 0.004), numerically reduced the Brier score from 0.106 to 0.102 ( P = 0.054), and retained acceptable calibration (intercept, -0.14; P = 0.144; slope, 0.89; P = 0.187). Higher SHR, interpreted in this clinically stable population as relative fasting glycemic elevation, was independently and consistently associated with single-visit albuminuria in two eastern-Chinese hospital health-examination cohorts. SHR provided a modest, externally reproducible same-visit risk-enrichment signal beyond routine clinical variables. It may help prioritize completion or repeat UACR assessment when urine testing is unavailable, declined, or requires confirmation, but it should not replace direct UACR measurement and does not establish acute stress hyperglycemia, kidney-disease diagnosis, causality, or longitudinal renal risk. Not applicable.