Aim: Deep vein thrombosis (DVT) is an insidious disease wherein more than 15 different clinical signs are described. The aim of this work was to focus on these clinical signs and to test them for their importance in making a diagnosis of DVT.
Methods: All patients treated with a tentative diagnosis of DVT in the emergency department were asked to take part in the study. Out of the 254 patients who were examined in order to exclude DVT, 204 patients agreed to participate in the study. The patients who agreed to take part were tested for fifteen clinical examination signs. The Wells score was then determined.
Results: Sixty-two were diagnosed with DVT. For 142 patients, DVT could be ruled out. The probability of DVT for 9 signs together is 88%, and for 3 signs is 78%. The negative predictive values are 91-95%. The combination of the clinical signs showed a specificity of 100%, independent if the patients were old, comorbid, and were diagnosed with the thrombosis in the lower limbs. The determination of the Wells score resulted in no convincing evidence for or against the diagnosis of DVT.
Conclusion: We suggest a modified Wells score integrating missing clinical signs with more reliable predictive values. Even with the availability of ultrasound, clinical signs have not become superfluous. They are quick to carry out, safe, cheap and an important addition to the Wells score, particularly for multimorbid and elderly patients.