We read with a great interest the review by Goodnough and Levy on the off-label use
of recombinant factor VIIa (rFVIIa), which was published under the title “The Judicious
Use of Recombinant Factor VIIa,” in an earlier issue of this journal.[1] The article clearly summarized the randomized trials with off-label rFVIIa performed
in the past 20 years. The association of off-label rFVIIa with uncertain efficacy
and mortality benefit and a significant risk of thromboembolic, especially arterial,
events were emphasized. The authors stated that rFVIIa should be used with restraint
in the setting of life-threatening hemorrhage and the risks of thromboembolic events
should be evaluated carefully and individually. Although we do agree that thromboembolic
events represent a serious complication of off-label rFVIIa, especially in a high-risk
population (elderly, cardiovascular comorbidity, cardiosurgery, and intracranial bleeding),
we would like to point out an issue that was not separately discussed in the review
and could, at least in our opinion, improve the safety of rFVIIa while maintaining
its efficacy.
The off-label studies and patient registries have reported the administration of rFVIIa
mainly in doses originally intended for the hemophilia population, for example, single
dose approximately 90 μg/kg or above. However, as shown in FVII deficiency, rFVIIa
could, at least in certain conditions, be hemostatically effective in substantially
lower doses. The experience with lower dosing in off-label setting remains limited,
although some of the randomized trials performed included arms with single doses as
low as 5 μg/kg.[2] Interestingly, most of those trials failed to show the superiority of arms with
rFVIIa ≥ 80 μg/kg in terms of clinical outcome (efficacy, mortality, demands on transfusion
therapy). The retrospective analysis by Schmid et al, that evaluated the intentional
off-label use of low-dose (56–71 µg/kg) rFVIIa in the setting of life-threatening
hemorrhage (blood loss ≥ 1,000 mL/h) in a single institution and included 73 cases
(49 related to surgery or trauma), found the comparable clinical outcome as reported
for the “standard” (∼90 μg/kg) rFVIIa doses. The rate of thromboembolic events in
the study was surprisingly low — only two (2.7%) patients were affected.[3]
We were inspired by the above-mentioned study and performed a retrospective analysis
(9-year period; unpublished data) of off-label rFVIIa use at our institution—a university
hospital with specialized units for hepatobiliary surgery and renal transplantation.
The analysis was intentionally restricted to the treatment of life-threatening bleeding
related to surgery, since rFVIIa was given particularly often in those patient subgroups.
Forty-six consecutively treated surgical patients (mean: 59.0 years; range: 31–83
years; 27 men; 30 [65.2%] with underlying malignancy; 21 [45.6%] with hepatic or pancreatic
resection; 6 [13.1%] with renal transplantation; 4 [8.7%] with the resection of retroperitoneal
tumor; 3 [6.5%] with vascular surgery; 3 [6.5%] with lung resection; 9 [19.6%] with
other major surgical procedures) with 47 bleeding episodes were included. Twenty-one
patients with 21 hemorrhages were treated with low-dose rFVIIa (mean: 26.4 μg/kg;
range: 11.5–64.5 μg/kg), whereas 46 patients with 46 hemorrhages were given rFVIIa
at a dose above 80 μg/kg (mean: 92.7 μg/kg; range: 82.5–109.0 μg/kg). The actual dose
of rFVIIa was individually determined by the consultant hematologist according to
the availability of the drug and the patient's condition. The low-dose group showed
a comparable patients' survival (90.4 vs. 80.7%, 48 hours after the first rFVIIa administration)
and blood product consumption (4.2 vs. 6.0 transfusion units [TU] of red packed cells,
3.7 vs. 4.5 TU of fresh frozen plasma, and 0.2 vs. 0.1 TU of platelet concentrates,
during a 48-hour period after the first rFVIIa administration) to the group with standard
dose rFVIIa. No significant differences in age, gender, factors influencing rFVIIa
efficacy (time to rFVIIa administration, body temperature, pH, coagulopathy, prior
transfusion therapy), repeated rFVIIa administration, and the use of other hemostatic
agents (antifibrinolytics, hemostyptics, vasoconstrictors) or procedures were seen
between the groups. The numbers of hepatic and pancreatic resection as well as renal
transplantation were proportional in both groups. Only two thrombotic events possibly
related to rFVIIa were observed, one in each group — deep vein thrombosis of the lower
extremity in the low-dose group and myocardial infarction in the other group. It is
important to emphasize, however, that the patient from the low-dose group had a high
prothrombotic risk separate from the rFVIIa administration (high-risk orthopedic surgery,
age older than 80 years, prolonged immobilization, and severe cardiovascular comorbidity).
Thus, in our experience, a low-dose rFVIIa dose appeared to be a reasonably effective
and safe modality for the off-label treatment of life-threatening hemorrhage related
to surgery. Our experience also stresses the importance of careful evaluation for
prothrombotic risk factors before the rFVIIa administration, as we also pointed out
in our previous work.[4] Of course, the results of our study are limited by several factors, most prominently
by the small number of evaluated patients, retrospective design, and compassionate
use of rFVIIa, and should be interpreted with caution. However, we do believe that
the use of lower rFVIIa doses could represent a risk-reducing approach in the off-label
setting, at least in case of surgery-related bleeding.
