We read with interest the recent review of Siragusa1 on D-dimer testing in the emergency setting for the management of acute venous thromboembolism (VTE). Thanks to its high negative predictive value, D-dimer represents an excellent first-line test in patients with suspected VTE, and combined with a low pre-test clinical probability of disease, a negative result of this assay can safely exclude VTE. However, as correctly outlined in the author’s conclusions, D-dimer accuracy should be carefully assessed in each hospital before implementing it as a screening test in the emergency management of VTE. Moreover, D-dimer should never be used without a correct pre-clinical probability assessment, based on standardised scores, such as those proposed by Wells or by the Geneva’s group, or on gestalt of experts in VTE. We would like to further discuss some common misuses and misinterpretations of D-dimer test in clinical practice. First, a simple Bayesian statistical rule is that post-test probability of a disease is influenced by the prevalence of the disease in a given population. Consequently, it is clear that D-dimer performs better in the exclusion of VTE if the pre-test clinical probability is low, thus in a population with a low prevalence of the disease. For this reason, D-dimer testing needs to be incorporated in diagnostic algorithms in which it is used in combination with a probability score, and only in a population of patients with a low or low-tomoderate pre-test clinical probability its negative result can be applied to safely rule out VTE2,3. Therefore, physicians should never exclude pulmonary embolism or deep venous thrombosis based on a negative Ddimer alone in a patient with a high-probability suspicion. Second, as correctly reported in the review, many physiological and pathological conditions can increase plasma D-dimer levels1. For this reason, and for the consequent low positive predictive value of the test, Ddimer should be only assessed in the presence of a precise clinical suspicion of VTE. It is a common malpractice in some emergency departments to use several laboratory tests (troponin, brain natriuretic peptide, and D-dimer) as generic screening markers of dyspnoea and/or chest pain4. Given the high likelihood of false positive results, physicians are then often forced to prescribe unnecessary additional radiological tests. Third, as a direct consequence of the previous point, elevated D-dimer levels should never be interpreted as a clear demonstration of an acute VTE. D-dimer test has been developed and tested as a screening assay for the exclusion of VTE (even if it is also used in the diagnosis of disseminated intravascular coagulation). Based on our current knowledge, elevated D-dimer levels cannot be considered as clinically relevant per se. Inappropriate detection of a positive test, in particular when performed in hospitalised patients, would require an unacceptably high number of subsequent additional diagnostic tests to rule in a very limited number of venous thromboembolic events. In conclusion, we believe that correct information on the best clinical use of D-dimer testing is necessary to improve its application in clinical practice. This would likely lead to a reduction in prescribed D-dimer tests and to a reduction in the so-called “D-dimer disease”, one of the clinically inexistent new laboratory diseases that have been created by the excessively increased use of blood tests.