Objective Patients with symptoms of deep vein thrombosis (DVT) and pulmonary embolism (PE) commonly present to the emergency department (ED). The aim of this study was to assess the role of ischaemia-modified albumin (IMA) testing in the diagnosis of venous thromboembolism (VTE).

Methods This was a prospective diagnostic cohort study. Inpatients and ED patients >16 years of age investigated for PE or DVT at a single hospital were eligible for study consent. Blinded IMA analysis was performed on the first blood sample taken from each patient. Patients underwent reference standard investigation for PE or DVT, including 3-month follow-up. Receiver operating characteristic (ROC) curves were constructed for IMA and the IMA:albumin ratio in the diagnosis of all VTE, PE and DVT. A sensitivity analysis was performed.

Results 452 patients were consented and investigated for DVT, and 354 patients were consented and investigated for PE (806 in total). 348 patients investigated for PE had IMA testing as did 195 of the first 199 DVT patients. VTE prevalence was 19.7%. The IMA:albumin ratio performed better than IMA alone. The area under the ROC curve (AUC) for IMA:albumin in all VTE was 0.60 (95% CI 0.54 to 0.66), in DVT 0.56 (95% CI 0.46 to 0.65) and in PE 0.63 (95% CI 0.56 to 0.71). In ED patients with symptoms of PE, the AUC for IMA:albumin was 0.69 (95% CI 0.60 to 0.78).

Conclusions IMA testing cannot be used alone to diagnose DVT or PE, although there is a moderate association with PE in ED patients.