Association between stress hyperglycaemia ratio and admission stroke severity: a retrospective study in a Chinese single-centre cohort.

Abstract

To evaluate the association between the stress hyperglycaemia ratio (SHR) and baseline stroke severity in patients with acute ischaemic stroke (AIS) and to investigate whether the relationship is non-linear. Retrospective cohort study. A tertiary hospital in Zhejiang Province, China. 1479 consecutive AIS patients admitted within 24 hours of symptom onset between 2016 and 2022. SHR was calculated as fasting plasma glucose (mmol/L) divided by glycated haemoglobin (HbA1c, %). Stroke severity was assessed by the NIH Stroke Scale (NIHSS) and categorised as mild (NIHSS ≤5) or moderate to severe (NIHSS >5). Associations between SHR and stroke severity were examined using multivariable logistic regression, generalised additive models and threshold effect analysis. Patients with more severe strokes had significantly higher SHR values (median 0.99 vs 0.94; p<0.001). After full adjustment, higher SHR remained independently associated with increased odds of moderate-to-severe stroke (OR=2.90; 95% CI: 1.60 to 5.26; p=0.0004). SHR in the highest quartile (Q4) was linked to a greater risk compared with the lowest quartile (Q1) (OR=1.90; 95% CI: 1.31 to 2.76; p=0.0008). Non-linear analysis revealed a threshold at SHR=1.3, with sharply increased risk below this point (OR=6.11; 95% CI: 2.64 to 14.15; p<0.0001). Subgroup findings were consistent, with a significant interaction across TOAST subtypes (P for interaction=0.0042). SHR is independently associated with greater stroke severity at admission. Values below 1.3 may reflect heightened metabolic stress and could help inform early risk stratification in AIS management, but their discriminative power is limited and should be interpreted in conjunction with other clinical indicators.