Current strategies for diagnosing pulmonary embolism (PE) include a clinical decision rule (CDR), followed by a D-dimer assay in patients with an unlikely clinical probability. We assessed the implementation of the current guidelines for the diagnosis of PE. A first questionnaire was sent to internists and pulmonologists to assess the proportion of physicians that adequately applied the guidelines. Two versions of a second questionnaire were sent presenting five hypothetical cases of which in two cases with an intermediate clinical probability an abnormal D-dimer test result was added to one version. We assessed the variation of the CDR and compared the proportions of a likely clinical probability between the two versions. A total of 65 physicians responded to the first questionnaire (response rate 75%). Half of the physicians (N=29; 46%) indicated that they use a CDR in all patients and 22 physicians (45%) indicated that they review the D-dimer result after they examined patients. Sixty-two physicians responded on the second questionnaire (response rate 36%). A shift was observed from an unlikely to a likely probability when an abnormal D-dimer test result was added to the clinical information (22% to 41%; p=0.22 and 26% to 50%; p<0.05). Our findings indicate that physicians do not use the guidelines for diagnosis of PE consistently. Furthermore, the knowledge of an abnormal D-dimer test result before seeing the patient leads to a higher CDR score. Physicians should therefore first examine patients before taking note of the D-dimer test result.