Determinants of futility of administration of recombinant factor VIIa in trauma.

BACKGROUND

"Off-label" use of human coagulation factor VIIa (FVIIa) is presently restricted to patients in extremis at our institution. Although bleeding will diminish in most patients, some will still die early as a result of irreversible shock and/or rebleeding. Futile administration of FVIIa significantly increases the economic burden of this expensive therapy and therefore limits its availability. On the basis of both human and in vitro studies, profound acidosis may be expected to predict lack of response. In addition, the depth of hemorrhagic shock, as defined by the degree of hypoperfusion over a given period of time, may be predictive of failure of FVIIa administration. We hypothesized that retrospective review of FVIIa use would identify variables associated with clinical futility.

METHODS

Characteristics of patients receiving FVIIa for acute traumatic hemorrhage were identified. Patients were retrospectively stratified into two groups; those who died as a result of acute hemorrhagic shock (nonresponders) and those in whom hemostasis was achieved and sustained (responders). Demographics, laboratory values, transfusion requirements, and outcomes were recorded for all patients. Data were analyzed using the Student's t test to identify the clinical characteristics of nonresponders and stepwise logistic regression was then used to identify independently predictive factors. A classification and regression tree analysis was conducted to develop a decision tree on the basis of our results.

RESULTS

Eighty-one patients received FVIIa therapy over a 3-year period. Among the 46 patients treated for acute hemorrhage, there were 26 with blunt and 20 with penetrating mechanisms of trauma. Average age was 35 +/- 15 years, 72% were male, and the average Injury Severity Score was 36 +/- 15. Revised Trauma Score (RTS), lactate, and preadministration prothrombin time (PT) each predicted lack of response (p < 0.05 for each). RTS and PT were independently predictive of failure of response. An RTS of less than 4.09 and a PT of greater than or equal to 17.6 seconds were significantly associated with futile administration of FVIIa. Age was a significant factor in patients with a PT greater than or equal to 17.6 seconds, whereas ISS was significant in patients with an RTS greater than or equal to 4.09.

CONCLUSION

Profound acidosis and coagulopathy may predict failure of FVIIa therapy. Depth of hemorrhagic shock, as described by the RTS, was also associated with futile administration. These variables should be considered as potential contraindications to the use of FVIIa. Earlier administration of FVIIa, before the development of massive blood loss and severe shock, may increase the rate of clinical response.