Combined use of the National Early Warning Score and D-dimer levels to predict 30-day and 365-day mortality in medical patients

Abstract

Aim: To investigate the combined use of NEWS and D-dimer levels to predict the 30-day and 365-day mortality rates of a cohort of Danish patients with complete follow-up.

Methods: Post-hoc retrospective observational study of acutely admitted medical patients aged 18 years or older who had D-dimer measured within 6h after arrival to two medical admission units in Denmark.

Results: The final study population consisted of 1201 patients with a median age of 65.0 years (range 18.0-107.0 years), and 44.7% were of male sex. Four patients (0.3%) died within 24h of admission, 69 (5.7%) within 30 days and 198 (16.5%) within 365 days. On admission, 576 (48%) patients had a NEWS≥3 - of these 441 had a D-dimer≥0.50mgL(-1): 55 (12.5%) of these patients died within 30 days, compared with 5 (3.7%) of the 135 patients with a D-dimer<0.50mgL(-1) (odds ratio 3.7, 95%CI 1.4-10.8). Nine of the 625 patients with a NEWS on admission <3 died within 30 days and all of these patients had a D-dimer≥0.50mgL(-1). None of the 218 patients with a D-dimer<0.50mgL(-1) died within 30 days of admission.

Conclusion: The combination of NEWS score<3 and D-dimer levels below 0.50mgL(-1) appears to identify patients of low risk of mortality within 30 days and, therefore, may prove to be a powerful risk assessment tool for acutely ill medical patients.

Keywords: D-dimer; Early warning scores; Emergency department; Mortality; Risk stratification; predictive scores.