THERAPY

Criteria abstracted from The Users' Guides to the Medical Literature series in JAMA

N Engl J Med. 2005 Feb 24;352(8):777-85.[abstract]

Review posted August 14, 2007

1. What is being studied?
   
   
   1. The study objective:
      
      The study aimed to evaluate the efficacy of activated recombinant factor VII (rFVIIa) in decreasing intracerebral hematoma (ICH) growth in the first twenty-four hours as compared to placebo in adults with hemorrhagic stroke.
      
      
   2. The study design:
      
      This was a multi-center, prospective, randomized, double-blinded, placebo-controlled trial. The patients included:
      
      
   3. Inclusion Criteria:
      
      Patients > 18 years of age were invited to participate in this study in that had a spontaneous intracerebral hemorrhage documented by a head CT scan within three hours of the onset of symptoms. Enrollment was conducted between August 2002 and March 2004.
      
      
   4. Exclusion Criteria:
      
      Patients with Glasgow Coma Scale < 5, patients who were to undergo surgical evacuation of hematoma within 24 hours of admission, and patients with anatomic or traumatic causes for ICH were all excluded Other risk factors leading to exclusion included patients at risk for abnormal clotting, such as those with coagulopathy, thrombocytopenia, anticoagulant use, and pregnancy, immediate history of symptomatic vaso-occlusive disease (<30 days), and preexisting neurologic disability secondary to a prior ICH.
      
      
   5. The interventions compared:
      
      Patients were randomly assigned to receive a single intravenous dose of placebo, 40, 90 or 160 mcg/kg of rFVIIa. Treatment was initiated within one hour after obtaining baseline head CT and no later then four hours after onset of symptoms.
      
      
   6. The outcomes evaluated:
      
      The primary outcome measured was the percent decrease in the size of the intracerebral hemorrhage on CT scan from baseline to 24 hours in the treatment groups, compared to the placebo group. Two neuroradiologists, who were blinded to treatment assignments, independently determined the volume of intracerebral hemorrhage.
      
      Secondary outcomes included mortality at 90 days, clinical status at 90 days, and number of serious thromboembolic events at 90 days. Objective scales and clinical exam were used to describe the degree of neurologic disability. Rankin Scale, Extended Glasgow Outcome Scale (EGOS), and Barthel Index were used to assess the global neurological outcome at 90 days.
      
      
2. Are the result of the study valid?
   
   Primary Endpoints:
   
   
   1. Was the assignment of patients to treatment randomized?
      
      Yes. Patients were randomized by sequentially numbered identical containers in blocks of four.
      
      
   2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?
      
      Of the four-hundred patients that underwent randomization, one patient subsequently withdrew consent. Of the remaining 399 patients, 396 patients had a baseline CT scan, and 384 (96% of the population) patients underwent CT scan at 24 hours.
      
      Was follow up complete?
      
      The Rankin Scale and EGOS values were not available for one patient in the 80- mcg/kg cohort. If a patient died or was discharged, the last available outcome scores were carried forward. In the various groups, the number of patients who had data carried forward from 15 days was 7, 6, 5 and 2 in the placebo, 40, 80, 160 mcg/kg groups, respectively
      
      Were the patients analyzed in the groups to which they were randomized?
      
      Yes. There were four arms to the study and each arm was compared to placebo. A comparison was also made between the combined rFVIIa group (all the patients that received rFVIIa) and placebo group. The analysis done in this study was conducted according to intention to treat principle.
      
      
   3. Were patients, health workers, and study personnel “blind” to treatment?
      
      Yes. Both the medical team and the patients were blinded to the treatment. The neuroradiologists reading the CT scans were blinded to treatment as well. It is unclear if the outcome assessors were blinded to treatment assignment.
      
      
   4. Were the groups similar at the start of the trial?
      
      Yes. Almost all characteristics of the population studied were similar. The groups were similar in terms of age, gender, ethnicity, and location of hemorrhage, initial GCS score, NIHSS score, initial systolic blood pressure, time to treatment, and mean volume of intracranial hemorrhage at baseline.
      
      
   5. Aside from the experimental intervention, were the groups treated equally?
      
      There was no stated difference in the medical management of the two groups. Although each institution was encouraged to medically manage the patient with ICH according to the 1999 American Heart Association Guidelines, it is not clear if the treatment groups received substantially different management apart from the study intervention.
      
      
3. What were the results?
   
   
   1. How large was the treatment effect and how precise was the estimate of the treatment effect?
      
      

      Primary Outcomes	Estimated mean relative increase (98.3% CI)	P value	Estimated mean absolute increase (98.3% CI)	P value
      Placebo	29 (16-44)	N/A	8.7 (4.9-12.4)	N/A
      rFVIIa 40 mcg/kg	16 (4-28)	0.07	5.4 (1.7-9.0)	0.13
      rFVIIa 80 mcg/kg	14 (2-27)	0.05	4.2 (0.3-8.0)	0.04
      rFVIIa 160 mcg/kg	11 (0-23)	0.02	2.9 (-0.8-6.6)	0.008

   * P value is indexed to placebo.
   
   The primary endpoint measured was degree of the mean percent increase in the volume of intracerebral hemorrhage in the first twenty-four hours. The mean absolute and relative percent increase in volume of the intracerebral hemorrhage was significantly lower in the groups given 80 and 160 mcg/kg when compared to placebo. This meant that there was a 5cc smaller increase in volume of the ICH with treatment of rFVIIa. Treatment with rFVIIa resulted in a relative reduction of 50% in the growth of the hematoma.
   
   In a subgroup analysis of 269 patients that received study drug within three hours after onset of symptoms, the treatment effect was even more pronounced. The mean increase in volume of the intracerebral hemorrhage was 34% for the placebo and with treatment it was 13% (P=0.004). The absolute increase in volume of intracerebral hemorrhage was 10.7 ml for the placebo group and 4.4 ml for the treatment group (P=0.009).
   

   Secondary Outcome: Mortality	Death/Total Group Patients (Percent)	Odds Ratio for survival (95% CI)	P Value
   Placebo	28  (29)	N/A
   rFVIIa 40mcg/kg	19 (18)	1.9 (1.0-3.8)	0.05
   rFVIIa 80 mcg/kg	17 (18)	1.8 (0.9-3.6)	0.10
   rFVIIa 160 mcg/kg	20 (19)	1.7 (0.9-3.3)	0.11
   Combined	56 (18)	1.8 (1.1-3.0)	0.02

The secondary outcomes sought to measure improvement in mortality. The 90-day mortality was significantly decreased with the pooled population treated with rFVIIa as compared to placebo. The 90-day mortality was 29% for placebo and 18 % for the combined treatment group. This represented a relative reduction of 38 percent (p=0.02 by the chi-square test but 0.10 by log-rank test). Interestingly, mortality improvement did not appear to correlate with the specific dose of rFVIIa that was given. Only at 40-micrograms per kilogram of rFVIIa and the pooled rFVIIa groups [OR 1.8 95%CI (1.1-3.0)] was a statistical improvement in mortality demonstrated relative to placebo. A closer inspection of the CI for the odds ratio shows a tight spread; however, the lower limit is less then one, but minimally less then one. If the OR for survival fell near the lower limit of the CI then the results may not be statistically significant given the sample size. However, if the true OR fell in the mid-point of the CI, then there would be a dramatic reduction in mortality.

Secondary Outcome: Clinical Status	EGOS Unfavorable outcome	OR for Improvement :(95% CI)	P value	NIHSS Median Score	P value	Barthel Index	P value	Modified Rankin Scale unfavorable outcome	OR for improvement  (95% CI)	P value
Placebo	78/96 (81)	N/A	N/A	12.5	N/A	25	N/A	66 (69)	N/A	N/A
rFVIIa 40mcg/kg	78/108 (72)	1.9 (0.9-3.8)	0.09	6.0	0.03	55	0.07	59 (55)	2.2 (1.1-4.0)	0.02
rFVIIa 80 mcg/kg	66/92 (72)	1.5 (0.7-3.2)	0.28	5.0	0.004	67.5	0.01	45 (49)	2.4 (1.3-4.6)	0.008
rFVIIa 160 mcg/kg	77/103	1.4 (0.7-3.0)	0.36	7.0	0.02	55	0.02	56 (54)	2.1 (1.1-4.1)	0.02

 

The global neurologic outcome at 90 days was assessed using 4 scales: the EGOS, NIHSS, Modified Rankin Scale, and the Barthel Index. Three of four tests that assess global outcome and neurologic deficit showed improvement with rFVIIa administration in a dose-related manner. The EGOS and NIHSS scales, used to measure extent of neurologic deficit, showed conflicting data. The EGOS scale did not show significant improvement in neurologic disability with administration of study drug, while the NIHSS scale showed significantly less impairment in neurologic status with study drug when compared to placebo. Using the Modified Rankin scale, scale used to assess global outcome, and the Barthel Index, used to analyze limitation in the activities of daily living, both showed statistically significant improvement. The confidence intervals for the odds ratio to assess the global outcome (MRS) are again narrow and do not fall below one. Even if the true OR fell near the lower spectrum, this would be statistically significant.

Secondary Outcome: Thromboembolic serial adverse events	Total (Percent)	Arterial	Venous
Placebo	2 (2)	0	2 (2)
rFVIIa 40mcg/kg	7 (6)	6 (6)	1 (1)
rFVIIa 80 mcg/kg	4 (4)	2 (2)	2 (2)
rFVIIa 160 mcg/kg	10 (10)	8 (8)	2 (2)

There were more thromboembolic events in the treatment group (N=21) as compared to placebo (N=2), but this was not statistically significant (P=0.12). However, statistical significance was achieved when arterial thromboembolic events are compared between treatment and placebo group. In the rFVIIa group, arterial thromboembolic events occurred in 5% of the patients (P=0.01) and in 0 patients in the placebo group. Of the complications that were due to arterial vasoocclusion, the two most common were myocardial infarction and cerebral infarction. Four of the nine cerebral infarcts were moderate or severe, and only two of theses events were thought not to be related to drug administration. Most of the myocardial complications were minor.

4. Will the results help me in carrying for my patients?
   
   
   1. Can the result be applied to my patient care?
      
      Most intracerebral hemorrhages in the pediatric patient are a result of structural abnormalities and a large part of the remaining intracerebral hemaorrhages are a consequence of an underlying disorder. Abnormal structure of the brain, prematurity, underlying coagulopathy, anatomic abnormality, thrombocytopenia, trauma, disseminated intravascular coaguloapthy, have all been implicated in the development of intracerebral hemorrhage. The common etiologies involved in the development of an intracerebral bleed in adults have been well studied but in pediatrics this has not been the case. This study excluded pediatric patients (age 0 years to 18 years). Therefore in the strictest sense it is not applicable to most of the patients cared for by pediatric intensivists. In the literature case reports exist which illustrate use of recombinant activated factor VII in achieving hemostasis in children with intracerebral bleeds when other aggressive medical management fails or is contraindicated1,2,3,4. Recently, the use of rFVIIa in achieving hemostasis has been described during cranial tumor resection when intensive medical therapy failed5. It may be useful to consider this as a treatment option in older patient whose cerebral hemorrhage was a result of head trauma but more trials are warranted. Future trials should incorporate neurologic outcome scoring that has been validated in the pediatric population.
      
      
   2. Were all clinically important outcomes considered?
      
      The investigators examined two key effects of rFVIIa in treating intraventricular hemorrhage, i.e., change in intracerebral hematoma volume and mortality. With administration of rFVIIa intraventricular hematoma volume increased significantly less than with placebo, but this would not be important if there was an increase in mortality or worse neurologic outcome with drug administration. There was improvement in neurologic outcome with administration of drug, but improvement in mortality was significant only at the lowest dose and with pooled data. The data suggest that more patients survived with better neurologic outcomes. Although there was an increase in thromboembolic events in the treatment group as compared to placebo group, this was not statistically significant, i.e., treatment with rFVIIa was not attributable to more thromboembolic events. In addition, occurrence of serious thromboembolic events associated with drug administration was reported. There were more thromboembolic events in the treatment group (N=21) as compared to placebo (N=2), but this was not statistically significant; however, statistical significance was achieved when arterial thromboembolic events are compared between treatment and placebo group. Of the complications that were due to arterial vasoocclusion the two most common were myocardial infarction and cerebral infarction.
      
      
   3. Are the likely treatment benefits worth the potential harms and cost?
      
      There are very few medical therapies for patients with cerebral hemorrhages. As such the mortality and morbidity rate for this diagnosis remains high. Children usually do not posses the same cardiac and cerebro-vacular co-morbidities that their adult counterparts do. Thus, the adverse events reported in this adult population may be less likely to occur among pediatric patients. Mortality decreased in the pooled population of rFVIIa and neurological improvement was seen in all patients that received rFVIIa as assessed by various rating scales, but it is not clear that the same benefits would be demonstrable among pediatric patients as the neurologic assessments have not been validated in pediatrics. In addition, the patient population selected for this study is very different from the pediatric patients that present with ICH. The potential benefits of rFVIIa in achieving hemostasis in ICH are undeniable, but the proper dose to be used is difficult to ascertain from the article. In the adult population, rFVIIa appears to be beneficial in light of existence of limited alternative therapy, but applying these findings to the pediatric population seems premature. Studies adequately powered to identify an outcome benefit in pediatric patients will need to be carried out. Cost associated with administration of rFVIIa was not addressed by this article. At our institution one dose (90mcg/kg) of rFVIIa for a 70 kg patient costs $4500 dollars. With the limited options available in treatment and high mortality rates, rFVIIa must be part of the armamentarium but should not be an automatic intervention.

References:

1. Brady, KM, et al., Recombinant activated factor VII (rFVIIa) treatment in infants with hemorrhage. Paediatr Anaesth. 2006 Oct;16(10):1042-6
   
   
2. Park, P, et al., Recombinant activated factor VII for the rapid correction of coagulopathy in nonhemophilic neurosurgical patients. Neurosurgery. 2003 Jul;53(1):34-8
   
   
3. Huang, WY, et. al, The use of recombinant activated factor VII in three patients with central nervous system hemorrhages associated with factor VII deficiency. Transfusion. 2004 Nov;44(11):1562-6
   
   
4. Morenski JD, et al., Recombinant activated factor VII for cerebral injury-induced coagulopathy in pediatric patients. Report of three cases and review of the literature. J Neurosurg. 2003 Mar;98(3):611-6.
   
   
5. Hartmann, et al., Recombinant activated factor VII in the treatment of near-fatal bleeding during pediatric brain tumor surgery. Report of two cases and review of the literature. J Neurosurg. 2006 Jan;104(1 Suppl):55-8
   
   
6. Basic and Clinical Biostatistics, Saunders and Trapp,copyright 1990 Appleton and Lange pages90-91
   
   
7. http://www.cche.net/usersguides/therapy.asp

Last Updated: August 14, 2007

Fill out our Feedback Form | Editorial Notes | About Us