We initiated a study to look at the consistency of the data using the thrombinoscope/Diagnostica Stago- Calibrated Automated Thrombogram™ (CAT) system. The CAT is a thrombin generation method described previously by Hemker. The goal of this phase of the study was to determine whether this test can be used as a clinical assay. First, we tried to determine the stability of the sample along with the normal reference range of the assay. For this, we collected samples from 20 normal healthy subjects to determine the stability of the sample and the normal reference range. The collected samples were tested at 0, 1 and 2 hours after collection. These samples had been kept at room temperature throughout the testing. Part of the aliquot samples were frozen at −80°C, and then the frozen sample tubes were thawed and analyzed at 2, 3, 4, and 8 weeks. The ETP determined in the 0, 1, and 2 hours after collection revealed no statistical significance in the mean values. This would indicate that a delay in starting the testing procedure in the freshly collected specimens up to 2 hours does not statistically affect the results. This would tend to indicate that blood samples drawn in the physician's office or satellite labs distally located from the core hospital labs may be suitable for testing if the specimen could be brought to the testing laboratory within 2 hours after the specimen has been centrifuged.
The plasma aliquot specimens that were frozen for 2, 3, 4, and 8 weeks were then thawed and run as described for the 0 hour plasma sample.
The 2 and 3 week specimens had statistically significant higher mean values compared to the mean values of the 0 hour mean value. However the 4 and 8 week values were not statistically different from the 0 hour run specimen. From these observations it appears that the process of freezing over the 2–3 week period, had an increased potential to form thrombin, but the longer freezing period of 4 to 8 weeks the data revealed no statistical difference from the 0 hour mean values.
Secondly, we tried to see if the reference ranges are different between the different populations. We have used the 4 week frozen sample ETP as our normal reference range (2149.3 nM/min +/− 455.2) in order to compare with other populations. For this, we performed the ETP from sample collected from patient with either factor V Leiden or prothrombin G20210A mutations, with a total of 15 patients included. The mean ETP was 2663.4 nM/min +/− 605, which was statistically different from the normal population (P<0.01). We also tried to see if the ETP from a lupus inhibitor (LI) patient is different from the normal reference range, with a total of 55 LI positive patients included. The mean ETP was 2284.6 nM/min +/− 539.8, which was not statistically significantly different from the normal reference range.
We also tried to determine the therapeutic range for the ETP in a patient on warfarin therapy. The patients in the study were LI negative, and neither the factor V Leiden nor the prothrombin G20210A mutations were detected. We grouped these patients into three categories: INR 1.3 to 1.9 (N=15), INR20–3.0(N=21) and INR>3.0 (N=8). The ETP values for these three groups were 1229.7 nM/min +/− 237.5, 916.6 nM/min +/−256.8 and 558.7 nM/min +/− 212.6. Based on our data, we postulated that the ETP between 650 – 1200 nM/min fall in the 2.0 – 3.0 INR range, which is the generally accepted therapeutic range.
In conclusion, the data indicate that the ETP assay can be used as a clinical assay. However, there is a need to establish the reference range for each of the patient populations under investigation. So far we only determined reference ranges for the LI patients and patients with inherited hypercoagulable disorders. We hope to establish the reference ranges for other subpopulations, such as pregnant women and hemophiliacs in the future.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.