Recombinant Factor VIIa for Bleeding in Non-hemophiliac Pediatric Patients

INTRODUCTION

Recombinant factor VIIa (rFVIIa) is indicated for treatment and prevention of bleeding in hemophilia patients with inhibitors to factors VIII or IX and in patients with congenital factor VII deficiency.1 Use of rFVIIa for other bleeding conditions has increased in recent years with demonstrated efficacy to control bleeding in a number of non-hemophilia coagulopathies in adults.2–4

Activated factor VII initiates coagulation tissue factor-dependent and independent pathways.5–7 After vascular wall or tissue injury, factor VII in plasma binds to exposed tissue factor (TF) and activates factors IX and X leading to thrombin generation and subsequently fibrin clot formation.589 Activated factor VII also directly activates factor X, independent of TF, by binding to activated platelets.10

Although less widely reported in pediatrics than adults, rFVIIa use for non-hemophiliac children with uncontrolled bleeding has been shown to be safe and effective.11 Reports in children describe a number of conditions that may benefit from rFVIIa including bleeding related to cardiac surgery, liver failure and transplantation, platelet disorders, neurosurgery, and disseminated intravascular coagulation (DIC).12–23 Current knowledge of coagulation function has helped broaden rFVIIa use for many of these indications, although data are still limited. The main objective of this study was to evaluate the use, efficacy, and safety of rFVIIa for the treatment of bleeding in non-hemophiliac pediatric patients at a single tertiary care pediatric center.

METHODS

A retrospective chart review was conducted of all patients receiving recombinant activated factor VII (rFVIIa, NovoSeven; NovoNordisk, Copenhagen, Denmark) from June 2003 through July 2005 at Riley Hospital for Children, Indianapolis. All patients < 18 years old were included in analysis. Patients with hemophilia A and B, as well as patients receiving rFVIIa for prophylaxis, were excluded. The hospital Institutional Review Board approved this study.

Demographic data collected included age, gender, weight, length of hospitalization, Intensive Care Unit length of stay, primary prescribing service, and presence or absence of a hematology consultation. Clinical data obtained included underlying indication for rFVIIa, transfusion of fresh frozen plasma (FFP), packed red blood cells (PRBC), platelets, cryoprecipitate, and concomitant medications influencing coagulation. Laboratory data included the prothrombin time (PT), international normalized ratio (INR), partial thromboplastin time (PTT), platelet count, fibrinogen level, factor VII activity, and d-dimer. All laboratory parameters from before and after (when available) rVIIa administration were recorded. Lastly, we recorded each individual dose (mcg/kg) and the total number or rFVIIa doses administered to each patient.

Indications for rFVIIa therapy were classified as bleeding from liver failure, DIC, gastrointestinal bleeding, post cardiac surgery, and other. The primary outcome of rFVIIa therapy was assessed through chart documentation with subsequent categorization of patients into one of the following five outcome groups: no change from baseline, bleeding resolution, additional rFVIIa dosing requirement with resolution, death associated with bleeding, and death from other causes. Primary outcomes are defined in Table 1.

Table 1. Primary outcome definitions

View Fullscreen

Mean dosing for outcome groups (resolution versus no resolution) were compared using independent samples t-test. The Kruskal-Wallis test was used to examine the differences of mean doses for each treatment indication. A paired-sample t-test was used to compare PT, INR, and PTT values before and after rFVIIa dosing. Statistical significance was assumed with P < 0.05. Data were analyzed using Statistical Package for Social Sciences (SPSS) software (version 14.0 Chicago, Illinois) and are presented as mean ± SD.

RESULTS

Forty-eight patients received rFVIIa during the study period of June 2003 through July 2005. Figure 1 illustrates study patients included and excluded from analysis. Twenty-four children ranging in age from 2 days to 17 years (median: 5.5 years) and weight from 1.2 to 88 kg (mean: 33.1 kg) were included in analysis, and received a total of 240 doses of rFVIIa.

View Figure

• Download as Powerpoint
• Download Figure

Treatment indications for rFVIIa included liver failure (n = 7, 29.2%), GI bleeding (n = 4, 16.7%), post cardiac surgery (n = 3, 12.5%), DIC (n = 3, 12.5%), and other (n = 7, 29.2%). Table 2 describes the characteristics of patients who received rFVIIa. Twenty-two (92.7%) patients received at least one transfusion of FFP, PRBC, platelets or cryoprecipitate.

Table 2. Characteristics of patients receiving recombinant factor VIIa therapy

View Fullscreen

The mean rFVIIa dose administered to all patients was 85.3 ± 20.1 mcg/kg. The mean doses used for the treatment indications of liver failure, GI bleeding, post-cardiac surgery, DIC, and others were 69.1, 88.9, 94.1, 89.9, and 89.5 mcg/kg, respectively, and were not significantly different (P = 0.475). However, the mean dose used for bleeding associated with liver failure (69.1 mcg/ kg) was 19% lower that the mean dose administered to all patients.

Bleeding resolution occurred in 13 (54%) patients after an average of 10.8 doses (range 1–46). Comparing responders to non-responders, neither the mean dose (81.9 vs. 86.7 mcg/kg, P = 0.4) nor the mean number of doses (10.8 vs. 9.1, P = 0.209) were significantly different. Bleeding non-responders were 50% younger (5.5 vs. 10.3 years, P = 0.104) than responders. Nine deaths occurred, but none were directly attributable to bleeding. Patient outcomes are listed in Table 2.

Pre and post rFVIIa dosing laboratory values for PT, INR, and PTT were available for 25 (10.4%) of the 240 total doses. The mean PT and INR before and after rFVIIa decreased from 17.9 to 14.0 seconds (P = 0.04) and from 1.6 to 1.2 (P = 0.05), respectively. The mean PTT increased after rFVIIa dosing from 45.1 to 48.7 seconds (P = 0.71). One adverse event was documented in the 24 patients. This patient experienced a right femoral deep vein thrombosis following rFVIIa therapy. The patient was subsequently started on a heparin infusion with no additional bleeding problems. No additional adverse events were documented.

DISCUSSION

Factor VIIa plays an important role in the coagulation cascade. Due to its mechanism of action, recombinant FVIIa has the potential to be an effective hemostatic agent for a number of bleeding conditions unrelated to hemophilia. There are many case reports and small case series describing rFVIIa for bleeding conditions in non-hemophilic children. Also, several retrospective studies have been conducted but have yielded data in a relatively small number of children.1315171824–27 Only three prospective studies have been performed.121416 Table 3 summarizes published data pertinent to our study regarding the use of rFVIIa for treatment of active bleeding. Recombinant factor VIIa administration resulted in bleeding resolution in 54% of our patients. The percentage of patients with a positive response to rFVIIa is slightly lower than that published in other studies; although our study is considerably larger than previous works.

Table 3. Summary of rFVIIa data for treatment of active bleeding non-hemophilic children

View Fullscreen

Table 3. Summary of rFVIIa data for treatment of active bleeding non-hemophilic children (cont.)

View Fullscreen

Pychynska-Pokorska and colleagues evaluated the efficacy of rFVIIa to control bleeding after cardiac surgery and cardiopulmonary bypass in eight children in a prospective open label study.14 Recombinant factor VIIa use reduced post-operative blood loss, and resulted in cessation of bleeding in 7 of 8 patients after doses of 30–60 mcg/ kg.14 Two retrospective studies demonstrated that rFVIIa significantly reduced blood loss post cardiac surgery after doses of 56–180 mcg/kg.1525 All of our rFVIIa recipients after cardiac surgery had resolution of bleeding. However, we did not quantify blood loss in these patients in the time periods surrounding rFVIIa dosing.

Bleeding associated with liver failure and GI bleeding accounted for 46% of the rFVIIa usage in our population. Our observed efficacy for these patients was less than in published reports that have shown bleeding resolution occurring in 71%–100% of patients.16172328–31 Brown and colleagues reported that 71% of patients with liver failure treated with rFVIIa (80 mcg/kg) for bleeding unresponsive to FFP and platelet transfusion had subjective improvement.17 The seven patients with liver failure in our study received an average rFVIIa dose of 69 mcg/kg. Two (29%) of our patients with liver failure had resolution of bleeding, four (57%) patients died unrelated to bleeding, and one had no clinical change from baseline. Reports of rFVIIa use in GI bleeding show that 50% of patients experience resolution of bleeding.232930 Doses in these reports varied from 4–270 mcg/kg. One report described use of a continuous infusion of rFVIIa following initial boluses.23 The average rFVIIa dose for GI bleeding in our patients was 88.9 mcg/kg, which is well within the range of other reports; however, no favorable responses were observed.

Older children tended to respond more favorably to rFVIIa dosing. The mean age of nonresponders (5.5 years) in our treatment group was about half that of responders (10.3 years). This difference could relate to the differences in pharmacokinetics of rFVIIa in children versus adults. Villar and colleagues demonstrated that after 90 mcg/kg doses, children had lower plasma levels of rFVIIa than adults.32 Children also have significantly faster clearance of rFVIIa than adults.3233 As the use of rFVIIa continues to increase in children without hemophilia, further pharmacokinetic and pharmacodynamic studies are needed to fully determine age-related differences.

Of concern was that one patient in our study developed a right femoral deep vein thrombosis after receiving rFVIIa. Only one other confirmed thrombotic adverse event attributable to rFVIIa has been reported in 138 children (0.7%).12–1825–27 The Food and Drug Administration's Adverse Event Reporting System (AERS) describes reports of 185 thromboembolic events following the use of rFVIIa.34 It is difficult to compare the AERS data with available safety data from pediatrics due to the lack of quality exposure data and the voluntary reporting nature of the AERS. However, a review of controlled clinical trials of rFVIIa showed that the incidence of thromboembolic events in rFVIIa treated patients (6%) did not differ from placebo treated patients (5.3%).35 Obviously, it is difficult to determine true incidence and risk of thromboembolic adverse events after rFVIIa given the limited number of controlled trials in children without hemophilia. Additional safety studies are warranted.

Formal guidelines for rFVIIa use at our institution were not in place during the study period. The rFVIIa indications and doses used in our study are consistent with previous reports, but that alone does not suggest that rFVIIa therapy was indicated or appropriate for each patient. Also, while this study was not designed to evaluate cost effectiveness, it is important to consider the drug-related expenditures along with the 54% efficacy. In the 24 patients in our study, drug costs for rFVIIa totaled approximately $797,000 with a median treatment course cost of approximately $10,700. Furthermore, 21 patients were primarily cared for by a service other than pediatric hematology, but only 43% of those had a hematology consultation to assist with the management of bleeding. Given the drug costs and overall bleeding resolution in our study, formal dosing guidelines that incorporate hematology consultation could be instituted to improve appropriateness and overall cost-efficiency.

Our sample size is relatively small, although it is one of the largest studies of recombinant activated factor VII in non-hemophiliac children. Additionally, our study was not designed to predict patient specific factors indicative of positive or negative response to rFVIIa. Therefore, we cannot conclude with certainty why 46% of our patients did not respond to rFVIIa, despite having received similar mean mcg/kg doses and mean quantity of doses as patients who did respond. Also, this study did not completely evaluate timing of rFVIIa administration in relation to the patient's clinical bleeding episode in order to correlate treatment outcome. Lastly, laboratory parameters such as PT, PTT, INR and fibrinogen levels were not consistently obtained, thus resulting in difficulties in analyzing normalization of laboratory values as an indicator of appropriate response.

CONCLUSIONS

This study presents additional data further describing the efficacy and safety of rFVIIa for bleeding unrelated to hemophilia in pediatric patients. Bleeding resolution for this population may be achieved using doses similar to those recommended for children with hemophilia. However, our results do not support the widespread use of rFVIIa for treatment of bleeding in non-hemophilia children. Hospital policies designed to guide practitioners towards appropriate use of rFVIIa should be considered. Based on the pharmacokinetics of recombinant factor VIIa and the observation that younger patients tended to not respond as well as older patients, additional research is needed to determine age-dependent variables that affect dosing requirements in non-hemophilic children. Prospective randomized controlled trials are needed to determine the role of rFVIIa in non-hemophilic children with significant bleeding and to further clarify its safety profile.