Infrared thermal imaging as a novel evaluation method for deep vein thrombosis in lower limbs

Early detection of deep vein thrombosis (DVT) is critical to prevent clinical pulmonary thromboembolism. However, most conventional methods for diagnosing DVT are functionally limited and complicated. The aim of this study was to evaluate the value of infrared-thermal-imaging (IRTI), a novel imaging detection or screening technique, in diagnosis of DVT in animal models.

DVT model of femoral veins was established in nine New Zealand rabbits. The right hind femoral vein was embolized and the contralateral one served as a nonembolized control. Measurements of IRTI, compression ultrasonography (CPUS), and angiography under ultrasonic observation (AGUO) were performed at three time points: T1 (baseline, 10 min prior to surgery), T2 (2 h after thrombin injection), and T3 (48 h postoperatively). Qualitative pseudocolor analysis and quantitative temperature analysis were performed based on mean area temperature (Tav) and mean curvilinear temperature (Tca) of the region of interest as shown in IRTI. Temperature differences (TD) in Tav (TD(Tav)) and Tca (TD(Tca)) between the DVT and control sides were computed. Comparative statistical analysis was carried out by paired t-test and repeated measure, while multiple comparisons were performed by using Greenhouse-Geisser and Bonferroni approach. Values of P < 0.05 and P < 0.01 were considered statistically significant and highly significant.

Modeling of DVT was successful in all rabbits, as confirmed by CPUS and AGUO and immediately detected by IRTI. IRTI qualitative analysis of pseudocolor revealed that the bilateral temperatures were apparently asymmetrical and that there were abnormally high temperature zones on the DVT side where thrombosis formed. The results of paired t-test of Tav and Tca between DVT side and control sides did not reveal statistical difference at T1 (Tav: P = 0.817; Tca: P = 0.983) yet showed statistical differences at both T2 (Tav: P = 0.023; Tca: P = 0.021) and T3 (Tav: P = 0.016; Tca: P = 0.028). Results of repeated measure and multiple comparisons of TD(Tav) and TD(Tca) were highly different and significant differences across the T2 (TD(Tav): P = 0.009; TD(Tav): P = 0.03) and T3 (TD(Tav): P = 0.015; TD(Tav): P = 0.021).

IRTI temperature quantitative analysis may help further detection of DVT. Additionally, IRTI could serve as a novel detection and screening tool for DVT due to its convenience, rapid response, and high sensitivity.