Thromboelastography (TEG®) is a potentially useful tool but analysis within 4-6 min of collection imposes limitations on its use and access. The use of citrate blood tubes potentially increases the time frame for processing specimens. There is, however, limited research on the stability of citrate specimens, timing of processing and the accuracy of TEG® results. The purpose of this study was to examine the effects of early and delayed processing on TEG® parameters using kaolinactivated citrated blood samples in the intensive care population. TEG® analysis was performed on 61 patients. Blood was collected into two 3.2% sodium citrate (0.105 m) tubes. Kaolin-activated samples were analysed at 15, 30 and 120 min postcollection. TEG® parameters analysed included reaction time (R), clot formation time (K), alpha angle (a), maximum amplitude, LY30, the coagulation index, time to maximum rate of thrombus generation, maximum rate of thrombus generation and total thrombus generation. Sixty-one critically ill patients were included. The results of the anova showed that time from collection was significantly associated with the TEG® results (P < 0.05). On comparison of individual outcome variables, this difference in most cases was due to changes over time from 30 to 120 min. Furthermore, progressive changes in TEG® parameters such as decreasing R were suggestive of a trend toward hypercoagulability of the specimens. Processing of kaolin-activated citrate TEG® specimens can begin as early as 15 min postvenipuncture. However, delaying processing by more than 30 min leads to a significant change in results.