Background Few data are available on the accuracy of thermodilution measurements of cardiac index with room-temperature injectates and a closed delivery system in patients with low cardiac indexes and/or hypothermic body temperatures. Objective To compare iced and room-temperature injectates for thermodilution measurement of cardiac index in postoperative cardiac surgical patients during hypothermia and normothermia. Methods In a convenience sample of cardiac surgical patients in a cardiac recovery unit, cardiac index was measured with both room-temperature and iced injectates during hypothermia (≤36.0ºC) and normothermia (≥36.1ºC and ≤38.0ºC). Device bias and precision were calculated and graphed by using the Bland-Altman method. A Student t test was used to determine differences between cardiac indexes by injectate temperature. Results A total of 38 patients were studied. Mean bias and precision for roomtemperature and iced injectates in all patients were 0.11 (SD, 0.27) during hypothermia and -0.03 (SD, 0.21) during normothermia. In hypothermic patients, car diac index differed significantly between room-temperature and iced injectates (t 1,37 = 2.41, P = .02). Cardiac index measurements did not differ between roomtemperature and iced injectates in normothermic patients (P = .33). Conclusions Although significant differences in thermodilution cardiac index were found between room-temperature and iced injectates during hypothermic body temperatures, these differences were small (mean, <0.11). These findings add to the results of the few studies on accuracy of room-temperature injectates for thermodilution measurement of cardiac index. (American Journal of Critical Care. 2010;19:365-372) Key directives in the evidence-based practice guidelines include (1) closed-system injection delivery systems should be used to decrease risk of infection, (2) either 5-mL or 10-mL injectate volumes can be used in patients with normal cardiac index, and (3) either room-temperature or iced injectates can be used in patients with normal cardiac index. 26The guidelines urge caution when measuring cardiac index in patients with low cardiac index, because few studies have been done in these patients to ensure the accuracy of cardiac index measurement with room-temperature injectates.The concern with the use of lowvolume or room-temperature injectates in patients with low cardiac index is that the signal-to-noise ratio may be decreased in these situations, leading to inaccurate measurements of cardiac index. Similar concerns arise in situations where the difference between injectate temperature and body temperature are less pronounced, as occurs in hypothermic body temperatures. 1,8,12 In a few studies, researchers have evaluated the accuracy of thermodilution measurements of cardiac index during hypothermic body temperatures 3,15 and/or in patients with a low cardiac index. 6,9,10,15 Of particular concern is that many of these studies did not use closed systems for injectate delivery, 3,10,15 and such systems are exclusively us...