Influence of C-reactive protein levels and age on the value of D-dimer in diagnosing pulmonary embolism

Background: Recently, the number of performed CT-angiographies to diagnose pulmonary embolism (PE) rised markedly, while the incidence of PE hardly increased. This low yield of CT-angiography leads to more patients exposed to radiation and higher costs.

Aim: The diagnostic value of age, C-reactive protein (CRP) and D-dimer in PE was investigated. Additionally an age-adjusted D-dimer cutoff level [age-adjusted cutoff = age/100 mg/L] was compared with the conventional cutoff level in diagnosing PE for patients ≥50 yr.

Methods: This observational study (2004-2007) included all consecutive patients suspected for PE presenting on the emergency department with a performed CT-angiography after measuring CRP and D-dimer levels.

Results: Of 4609 patients suspected for PE, 1164 patients underwent CT-angiography of whom 309 (26.5%) had PE. Correlation between CRP and D-dimer was 0.42 (P < 0.001). D-dimer and age correlated positively (rs = 0.33, P < 0.001), but only in patients >50 yr and independent of PE. Multivariate regression analysis showed significant contribution of age, D-dimer and age-adjusted D-dimer for diagnosing PE, but not for CRP. Using an age-adjusted D-dimer cutoff value increased specificity from 37% to 50%, whereas sensitivity declined from 96% to 90%. Applying this age-adjusted cutoff level in patients ≥70 yr, specificity increased from 18% to 40%, while sensitivity decreased from 96% to 88%.

Conclusions: In the prediction of PE, age and D-dimer levels are relevant, while CRP level is not. Using an age-adjusted D-dimer cutoff in older patients remarkably improves the specificity of D-dimer testing with a minor decline in sensitivity. This may increase the yield of CT-angiography in diagnosing PE.