Analyzing the correlation between the longitudinal trajectories of blood urea nitrogen and 30-day survival in hemorrhagic stroke patients.

Background: While some studies suggest a link between blood urea nitrogen (BUN) levels and adverse outcomes in hemorrhagic stroke (HS) patients, the prognostic value of longitudinal BUN changes remains unclear.

Objective: To evaluate the association between longitudinal BUN trajectories and 30-day mortality risk in HS patients.

Methods: We analyzed HS patients from the MIMIC-IV database diagnosed within 24 hours of hospitalization. Group-based trajectory modeling (GBTM) was used to identify BUN trajectories. Kaplan-Meier survival curves and Cox proportional hazards models were employed to assess mortality risk, while ROC curves evaluated BUN's predictive accuracy.

Results: Among 1,172 HS patients, three distinct BUN trajectories were identified. Patients with rising BUN trends (Class 2 and 3) had significantly higher 30-day mortality risks compared to those with stable BUN levels (Class 1) (HR > 1, p < 0.001), with Class 3 patients exhibiting the worst outcomes. ROC analysis demonstrated strong predictive accuracy for mortality, with AUC values of 0.866, 0.841, and 0.841 at 7, 14, and 30 days, respectively, after adjusting for confounders.

Conclusion: Persistently elevated BUN trajectories are independently associated with increased 30-day mortality in HS patients. This study highlights the heterogeneity of BUN trajectories in HS, providing insights beyond baseline BUN measurements and enhancing understanding of HS progression.