Patients with joint replacement were generally older. If the traditional D-dimer value was used as a preoperative diagnosis method, the false positive patient would be greatly increased. [5]
We used both age-adjusted D-dimer and traditional D-dimer valuesto detect the effectiveness of deep vein thrombosis. The area under the ROC curve was 0.795 and 0.685, respectively, and the age-adjusted D-dimer had a better predictive value in predicting deep vein thrombosis. In addition, we found that the sensitivity of the age-adjusted D-dimer cut-off value (76.8%) was significantly improved compared to the traditional method (41%), while the specificity, age-adjusted D-dimer cut-off value (76.9%) was also slightly better than the traditional D-dimer (65.8%).
The basic D-dimer value increased with age. P. L. Harper et al. found that D-dimer concentration increased with age reducing the clinical value of the D-dimer assay in the elderly. [5]
Through the age-stratification study, we found that the elderly had higher incidence of thrombosis, which is consistent with the conclusion of our risk factor analysis, and by the age-adjusted D-dimer cut-off value in all age groups, it can be found that more samples are lower than the cut-off value, which improved the sensitivity of the model.
In our study, the age-adjusted D-dimer value was defined as age x 0.01 (µg/mL), which was considered safe. Mads Nybo et al. had found that the age-adjusted D-dimer cut-off was safe to recommend the use of an age-adjusted D-dimer in a DVT setting as well as for PE from a systematic review. [9] In this study, we also made multivariate logistic regression analysis to find that age is the risk factor of DVT.
In previous reports, Norio Imai et al. demonstrated that age and D-dimer index can be useful in screening patients for DVTs before THA. [13] The conclusion was similar to us. Moreover, age-adjusted D-dimer value was calculated as patients' age times 0.01(µg/mL) by Henrike J Schouten et al. [10] They drew conclusions that age-adjusted D-dimer value can increase specificity, whereas sensitivity was not. The sensitivity was 97%, but the specificity was below 60%. The results did not coincide with our research. In addition, some studies focused on the performance of age adjusted D-dimer cut-off values for reducing the use of ultrasounds. [3] [14] [15] Kelly Broen et al. inclined the view that age adjusted D-dimer cut-off values can lead to less duplex ultrasounds performed to find patients that suffers from DVT. [16]
In our study, the age-adjusted D-dimer value cut-off increased sensitivity, which was important to find patients with DVT. DVT was the most critical complication. Once we did not observe the asymptomatic DVT patients, it might cause serious consequences, such as pulmonary embolism (PE). Thus, the age-adjusted D-dimer could help us find more asymptomatic DVT patients. Actually, that was what we cared about more.
As far as we know, it was the first study to focus on the performance of age adjusted D-dimer cut-off values for the screening of deep vein thrombosis in both TKA and THA. We gave the lower extremity venous color ultrasound bedside to all patients admitted to the hospital. Thus, we had enough samples for observation.
However, there were several limitations in our study. First, we did not use Wills score to assess the sensitivity and specificity of age-adjusted D-dimer cut-off value. Wills score could help us predict DVT better. [17] Second, our hospital used latex agglutination turbidimetry (LATEX) to measure the concentrations of D-dimer. Compared to enzyme linked fluorescent assays (ELISA), LATEX may lead lower sensitivity and specificity in our study. Finally, we did not group patients to several level by age, which may get a better result in this study.
In conclusion, compared to traditional D-dimer, age-adjusted D-dimer showed better performance in screening DVT, which was useful clinically. Meanwhile, more clinical observations were needed to verify our ideas.