Comparison of infrared camera measurement and color doppler ultrasonography performed by emergency physicians in diagnosis of deep venous thrombosis

Aim: To analyze the effectiveness of Wells score, D-dimer testing, infrared camera (IR) measurement and color Doppler ultrasonography (CUS), as performed by emergency physicians and radiologists. Materials and Methods: This prospective study was conducted in 71 patients with suspected deep vein thrombosis (DVT) and healthy volunteers of 60 individuals. Age, gender, comorbidities, medication use, fever, arterial blood pressure, pulse, peripheral oxygen saturation and Wells scores were recorded for the patients presenting with symptoms of DVT. IR imaging and CUS were performed by four trained emergency residents, and CUS was repeated using radiology. SPSS 18.0 was used for statistical analysis. Findings: Of the 71 patients, 15 (21.1%) had acute DVT, seven (9.9%) had subacute DVT, and four (5.6%) had chronic DVT. Cellulitis was detected in 13 patients (18.3%). The average temperature difference between both legs of the individuals in the control group was 0,17 °C and was statistically insignificant (p = 0.289). However, a statistically significant (p=0.04) leg temperature difference of 0.72 °C was observed in the experimental group. The sensitivity and specificity of each diagnostic method used in the study was 88.5%; 35.6% for IR camera, 100%; 95.5% for emergency bedside CUS, 96.2%; 24.4% for D-dimer, and 96.2%; 6.7% for Wells score. A significant thermal difference was detected in all patients with acute DVT, via IR camera and an artificial intelligence program. No thermal differences were found in one of seven patients with subacute DVT and two of four patients with chronic DVT. Only two patients would have received unnecessary anticoagulant treatment if all patients had been examined with emergency bedside CUS alone. If the algorithm formed by the combination of IR camera, CUS, and D-dimer had been used, two patients would have received unnecessary anticoagulant treatment and we would have missed one case of DVT. Results: Emergency physician-performed CUS is associated with high rates of accuracy, and is a cost-effective and less time-consuming diagnostic method with regard to lower extremity DVT. In addition, negative aspects, such as treatment delays and long waiting times in the emergency department, can be improved. IR camera measurement can be used as a pre-test, especially with regard to acute DVT. Key Words: Deep Venous Thrombosis, Venous Thromboembolism, Color Doppler Ultrasonography, Infrared Imagining, Artificial Intelligence, Emergency Department