Thrombophilia screening, including protein C and S determinations is an important part of the diagnostic investigation for idiopathic VTE in order to determine etiology, duration of anticoagulation, and to guide family testing [7, 8, 15]. It is widely believed that assays for natural coagulation inhibitors cannot be performed during the acute phase of VTE due to a consumptive decrease in protein C and S leading to false positive testing. This assumption is based on the study by D'Angelo et al that identified a mean decrease in protein S activity in patients with acute VTE before warfarin therapy. This transient state of acquired protein S deficiency was thought to be a result of increase in C4bBP levels secondary to inflammation. The study only examined 8 patients and provided only a mean protein S level rather than showing the proportion of patients that had a level below the reference range. The authors of this study commented that these results would need to be repeated before concluding that acute VTE falsely lowers levels of protein S. These results have not been replicated in any further studies and, to our knowledge, the only other studies investigating this issue are our previous study in 2006 and this follow up study.
Our previous study in 254 patients with acute VTE showed that 91% of patients had normal protein C and S levels and among those with abnormal initial tests 4/12 and 6/12 were proven to have PC and PS deficiency at a later date. This figure resulted in a sensitivity of 100% and specificity of 97% and 98% respectively with the caveat that patients with an initial normal test were not retested at a later date therefore we could not be sure of possible missed false negatives. The current study shows that our previous assumptions are correct and PC and PS testing can be done at the time of an acute VTE event with a high degree of specificity and sensitivity. Assuming a pre-test probability of 2% the negative likelihood ratio for PC testing is 0 (95% CI, 0%-3%) and for PS testing is 0 (95% CI, 0%-2%).
There is no indication from this series of patients that acute VTE alters the levels of protein C or S enough to affect the diagnosis of coagulation factor deficiency. Whereas genetic tests such as Factor V Leiden and the G20210A prothrombin gene variant can be tested at any time, coagulometric assays (such as protein C and S) cannot be performed in patients anticoagulated with vitamin K antagonists. Therefore, patients with normal protein C or S levels at the time of diagnosis can be effectively ruled out as having deficiency of these proteins and can therefore avoid the expense and inconvenience of repeat testing during VKA therapy interruption.
In conclusion, our current study in combination with our previous results show that testing protein C and S levels during the acute diagnosis of VTE is a valid approach. We suggest that thrombophilia testing be performed at the time of diagnosis of VTE, before VKA therapy is initiated. This is the most convenient time for testing and allows early identification of patients who may require more prolonged therapy or further screening of family members. Only those who have an initially abnormal test should have confirmatory testing repeated after a course of VKA therapy and may require LMWH bridging. In this study we did not assess whether other coagulation based assays, such as antithrombin, can be accurately tested at the time of acute VTE diagnosis. This would be an interesting avenue for future research. In any case, since warfarin does not affect antithrombin levels, testing for antithrombin deficiency is possible at a later date without the need for interrupting anticoagulation.