From bench to triage: diagnostic utility of delta-like canonical notch ligand-1 (DLL-1) for early sepsis prediction in the emergency department.

BACKGROUND

The early identification of patients with sepsis among those presenting with suspected infection in the emergency department (ED) remains challenging. Delta-like canonical Notch ligand-1 (DLL-1), a Notch pathway ligand involved in immune and endothelial signaling, has been shown to reflect disease severity and outcomes in ICU-based sepsis cohorts. Its value for early risk stratification at ED presentation is less well defined.

METHODS

We analyzed DLL-1 concentrations in serum samples from a prospectively enrolled ED cohort of adults presenting with clinically suspected acute infection. Blood samples were obtained during the initial clinical assessment. Sepsis was defined according to Sepsis-3 criteria and adjudicated by an expert panel, including evaluation of organ dysfunction within 72 h after presentation. DLL-1 concentrations were measured using enzyme-linked immunosorbent assay (ELISA), and diagnostic performance was assessed using AUROC analysis.

RESULTS

Among 300 enrolled patients, 74 (24.7%) were classified as having sepsis. DLL-1 concentrations were significantly higher in patients with sepsis than in those without sepsis (median 11,454 vs. 8,085 pg/mL; p < 0.001). DLL-1 demonstrated moderate discriminatory performance for sepsis at ED presentation (AUROC 0.69, 95% CI 0.62-0.76). Performance was lower than that of procalcitonin and NEWS2 but comparable to C-reactive protein and lactate. DLL-1 showed a moderate correlation with SOFA score. Importantly, correlations with CRP and white blood cell count were nonsignificant.

CONCLUSIONS

When measured at ED presentation, DLL-1 concentrations are elevated in patients with sepsis but showed only moderate diagnostic performance. These findings are consistent with earlier disease stages and lower degrees of organ dysfunction compared with ICU populations. DLL-1 may offer complementary host-response information beyond established biomarkers, but does not support stand-alone use for early sepsis diagnosis and warrants further evaluation within multimodal risk stratification strategies.

TRIAL REGISTRATION: DRKS00017395.