Role of a New Age-adjusted D-dimer Cutoff Value for Preoperative Deep Venous Thrombosis Exclusion in Elderly Patients With Hip Fractures

Abstract

Objective: This study aimed to describe the characteristics of plasma D-dimer level with increasing age and establish a new age-adjusted D-dimer cutoff value for excluding preoperative lower limb deep vein thrombosis (DVT) in elderly patients with hip fractures.Methods: This was a retrospective study of elderly patients who presented with acute hip fracture in our institution between June 2014 and May 2020. All patients underwent D-dimer test and duplex ultrasound. Patients were divided into six 5-year-apart age groups. The optimal cutoff value for each group was calculated by using receiver operating characteristic (ROC) curves, whereby the new age-adjusted D-dimer cutoff value was determined. The sensitivity, specificity, positive predictive values (PPV) and negative predictive values (NPV) were calculated and compared when different D-dimer cutoff values were applied, i.e. conventional 0.5mg/L, previously well-established age-adjusted cutoff value (age*0.01 mg/L) and the new age-adjusted D-dimer cutoff value herein. Results: There were 2759 included, 887 males and 1872 females, with an average age of 77.8 years. 280 patients were diagnosed with preoperative DVT. The optimal cutoff values for the six age groups were 0.715 mg/L, 1.170 mg/L, 1.620 mg/L, 1.665 mg/L, 1.690 mg/L and 1.985 mg/L, respectively and the calculated age-adjusted coefficient was 0.02 mg/L. With this new coefficient applied, the specificity was 61.3%, clearly higher than those for conventional threshold (0.5mg/L, 36.9%) or previously established age-adjusted D-dimer threshold (age*0.01 mg/L, 22.2%). In contrast, the sensitivity was lower than that (58.9% vs 85.0% or 77.1%) when D-dimer threshold of 0.5 mg/L or age-adjusted cutoff value (age*0.01 mg/L) was used. The other indexes as PPV (14.7%, 11.0% and 12.1%) and NPV (93.0%, 92.9% and 93.5%) were comparable when three different D-dimer thresholds were applied. Conclusions: The use of the new age-adjusted D-dimer threshold increased the proportion of elderly hip fracture patients in whom preoperative lower limb deep vein thrombosis could be safely excluded.