Background/Aim: Recirculation measured by thermodilution includes effects caused by access and cardiopulmonary recirculation. The aims of this study were to illustrate the accuracy of thermodilution in measurement of hemodialysis recirculation and also to identify a sensitive and specific threshold to detect access recirculation. Methods: 110 studies were performed in 19 patients. Recirculation obtained directly by the blood temperature monitor (BTM) was compared to that calculated from access blood flow, pump blood flow, and cardiac output determined by ultrasound dilution using the hemodialysis monitor (HDM). Results: A highly significant linear correlation was obtained between repeated BTM recirculation measurements (RBTM, 2 = 0.99·RBTM, 1 – 0.22%, r2 = 0.99). There were no significant differences between repeated BTM recirculation measurements with correct placement (11.4 ± 7.1 vs. 10.9 ± 7.4%, p = NS) or reversed placement (30.0 ± 15.6 vs. 30.2 ± 15.9%, p = NS) of blood lines. A strong linear relationship was obtained between the recirculation determined by thermodilution and the recirculation calculated from HDM measurements (Rcalc = 0.98·RBTM – 1.49%, r2 = 0.95). The mean recirculation obtained by BTM was not significantly different from the recirculation calculated by HDM with correct placement (9.5 ± 2.2 vs. 8.6 ± 2.5%, p = NS) or with reversed placement (25.4 ± 7.8 vs. 23.8 ± 7.7%, p = NS) of blood lines. When a recirculation greater than 15% measured by the BTM was considered as the threshold at which true access recirculation occurred, sensitivity and specificity of the thermodilution method to detect access recirculation were 93 and 98%, respectively. Conclusions: Recirculation measurements made by the BTM are accurate and precise. Even though BTM thermodilution includes effects of cardiopulmonary recirculation, so that low levels of access recirculation might not be detected, a BTM recirculation >15% represents a highly significant access recirculation.