A nomogram incorporating baseline NIHSS, admission HbA1c, 24-h capillary blood glucose standard deviation, fibrinogen, hs-CRP, and baseline DWI-ASPECTS demonstrated excellent discrimination (AUC 0.815) for predicting early neurological deterioration in acute ischemic stroke patients with type 2 diabetes mellitus, with 24-h CBG-SD emerging as the strongest predictor.
Early neurological deterioration occurred in 15.8% of AIS patients with T2DM within 72 hours of admission.
Six independent predictors of END were identified through LASSO and multivariable logistic regression analysis.
The nomogram demonstrated excellent discrimination for predicting END with an AUC of 0.815.
24-hour capillary blood glucose standard deviation (24-h CBG-SD) emerged as the strongest individual predictor of END.
Integrating metabolic and thrombo-inflammatory markers provided significant added predictive value over basic clinical features alone.
The study population consisted of 965 AIS patients with T2DM in a retrospective design.
Fibrinogen and hs-CRP, representing thrombo-inflammatory status, were independent predictors of END in AIS patients with T2DM.
Admission HbA1c was identified as an independent predictor of END, reflecting chronic glycemic control prior to the stroke event.
This research suggests that among people who have an acute ischemic stroke (a stroke caused by a blood clot) and also have type 2 diabetes, about 1 in 6 will experience early neurological deterioration — meaning their symptoms worsen significantly within the first three days. Researchers developed a prediction tool called a nomogram using data from 965 patients to identify who is at highest risk for this dangerous complication. The tool combines six factors: stroke severity on admission, a measure of long-term blood sugar control (HbA1c), how much blood sugar fluctuates in the first 24 hours, two markers of inflammation and clotting in the blood (hs-CRP and fibrinogen), and a brain imaging score. The tool performed well, correctly distinguishing between patients who did and did not deteriorate with an accuracy measure (AUC) of 0.815, and this held up in internal testing. A particularly notable finding is that rapid swings in blood sugar during the first 24 hours of hospitalization — measured as the standard deviation of bedside glucose readings — was the single strongest predictor of early worsening. This suggests that not just average blood sugar levels, but how much they fluctuate acutely, plays a critical role in stroke outcomes for diabetic patients. Inflammation and clotting markers also contributed meaningfully to risk prediction, pointing to a combined 'thrombo-inflammatory' process as important in early stroke worsening. This research suggests that carefully monitoring and potentially stabilizing blood sugar fluctuations in the first day of hospitalization, alongside tracking inflammatory and clotting markers, could help clinicians identify diabetic stroke patients who need the most intensive care. The nomogram provides a practical, validated tool that doctors could use at the bedside to calculate individualized risk scores, which could inform decisions about monitoring intensity and treatment strategies in this high-risk patient group.
Zhong X, Chen Z, Cao T, Fan C, Fu W, Jiang Y, et al.. (2026). Development and validation of a nomogram to predict early neurological deterioration in patients with acute ischemic stroke and type 2 diabetes mellitus: the pivotal role of glycemic variability and thrombo-inflammation.. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology. https://doi.org/10.1007/s10072-026-09160-8