Neurocritical Care

, Volume 10, Issue 1, pp 55–60

Perioperative Use of Recombinant Factor VII to Prevent Intraoperative Aneurysm Rupture in High Risk Patients: A Preliminary Safety Evaluation

Authors

    • National Brain Aneurysm CenterSt. Joseph’s Hospital
  • Tariq M. Janjua
    • National Brain Aneurysm CenterSt. Joseph’s Hospital
  • Archie Defillo
    • National Brain Aneurysm CenterSt. Joseph’s Hospital
  • Penny Sinner
    • National Brain Aneurysm CenterSt. Joseph’s Hospital
  • Andrea Zelensky
    • National Brain Aneurysm CenterSt. Joseph’s Hospital
ORIGINAL ARTICLE

DOI: 10.1007/s12028-008-9149-9

Cite this article as:
Nussbaum, E.S., Janjua, T.M., Defillo, A. et al. Neurocrit Care (2009) 10: 55. doi:10.1007/s12028-008-9149-9

Abstract

Object

The authors present a safety evaluation of the perioperative use of recombinant activated factor VII (rFVIIa) in a series of subarachnoid hemorrhage patients undergoing microsurgical aneurysm clipping.

Methods

We performed a retrospective chart review of the records of 18 consecutive subarachnoid hemorrhage patients who underwent craniotomy for aneurysm clipping and received an intraoperative dose of rFVIIa. In each case, the aneurysm was felt to be a “high risk” lesion for intraoperative rupture either because it had bled multiple times prior to surgery or based on anatomical considerations. All complications were recorded whether or not they were attributed to the use of rFVIIa.

Results

Eighteen patients, 7 men and 11 women, ranging in age from 42 to 85 years were included in this review. Nine patients (50%) were either Hunt/Hess Grades IV or V. Six patients developed clinically significant cerebral vasospasm. Fifteen patients required ventricular drainage on admission, and seven patients ultimately required a ventriculoperitoneal shunt. One lower extremity deep venous thrombosis and seven upper extremity venous thromboses in association with peripherally inserted central catheter (PICC) lines were identified on screening Doppler evaluations. There were no associated pulmonary emboli. No aneurysm ruptured intraoperatively.

Conclusion

We describe our experience with the use of a single dose of rFVIIa administered perioperatively in an attempt to decrease the rate of intraoperative aneurysm rupture in patients undergoing microsurgical aneurysm clipping. There were no significant immediate or long-term adverse effects attributable to the use of rFVIIa in this group, and we encountered no intraoperative ruptures in this relatively small series.

Keywords

AneurysmBrainFactor VIIRuptureSubarachnoid hemorrhage

Introduction

Intraoperative aneurysm rupture has been associated with substantial morbidity and mortality in patients undergoing microsurgical aneurysm clipping [7, 17, 19, 38]. Although the use of temporary vascular occlusion or intraoperative hypotension may decrease the risk of intraoperative rupture, some aneurysms will still rupture prematurely prior to completion of the dissection and well before the surgeon is prepared to apply the final clip. In these instances, significant morbidity may arise from attempts to stop the bleeding without the advantage of a bloodless operative field.

Recombinant activated factor VII (rFVIIa) is a relatively new and highly potent antifibrinolytic agent developed as a substitute for purified donor factor VII [11]. It has been used successfully to limit expansion of intracerebral hematomas and in the setting of severe trauma [2, 4, 23, 25, 31, 35]. We hypothesized that the administration of rFVIIa at the start of craniotomy could decrease the incidence of intraoperative aneurysm rupture without producing unfavorable thrombotic side effects. This study represents an initial experience with rFVIIa in patients undergoing microsurgical aneurysm clipping who were felt to be at high risk for intraoperative rupture.

Patients and Methods

Eighteen consecutive patients presenting with SAH who were felt to be at high risk for intraoperative aneurysm rupture were evaluated by retrospective chart review. We included patients who had suffered multiple hemorrhages prior to surgery, patients with large or complex aneurysms that were anticipated to require extensive dissection, patients with presumed “neck bleeds” based on aneurysm morphology (for example, the presence of a nipple at the aneurysm neck), and patients in whom it was felt that it would be difficult to achieve proximal control prior to aneurysm exposure. The rFVIIa was administered at the start of surgery as either a 2.4 mg or 4.8 mg IV bolus given over 2–5 min. The dosage was based on the patient weight and the manner in which the drug is commercially supplied (in dosage vials of either 2.4 mg or 4.8 mg), with 2.4 mg used in patients under 60 kg and 4.8 mg in those over 60 kg. This dosage was within the range published in the literature: 40–160 µg/kg.

Patients were classified preoperatively based on the Hunt-Hess grading scheme. All complications were recorded whether or not they were felt to be related to the use of rFVIIa. The study was approved by the human subjects committee of the Saint Joseph’s Hospital Institutional Review Board.

All patients underwent daily transcranial Doppler ultrasound evaluation to monitor for cerebral vasospasm, and all underwent routine upper and lower extremity screening ultrasound evaluations twice weekly to monitor for venous thrombosis. All patients were treated with prophylactic sequential compression stockings which were started at the time of admission. Subcutaneous heparin therapy was typically initiated 48 h postoperatively as well. All patients received prophylactic colloidal volume expansion and oral nimodipine or intravenous nicardipine therapy.

Of note, at our center, all patients with intracranial aneurysms are evaluated by a team that includes a neurovascular surgeon and an interventional neuroradiologist. Patients with ruptured aneurysms, and in particular those in poor condition, are generally treated endovascularly if the aneurysm is suitable anatomically for coiling. The majority of patients treated with open microsurgery have complex and wide-necked lesions not readily amenable to coil embolization. From January through December, 2007, our service treated 350 intracranial aneurysms. This included 170 ruptured lesions. Of these, 108 were coiled and 62 were clipped, including the 18 patients in the present study. There were 180 unruptured aneurysms; 112 were clipped, and 68 were coiled.

IRB approval was obtained for retrospective chart review prior to data collection in this series. The majority of our patients had suffered multiple hemorrhages prior to arrival and underwent surgery on an emergent basis, including some with preoperative coagulopathy. In these cases, the decision to administer rFVIIa was based on our very favorable experience with the use of this agent in the setting of intracerebral hematoma. In each case, the use of rFVIIa was discussed with the patient’s family, preoperatively. The risks and potential complications associated with the use of the drug were described. Our experience with similar “high risk” aneurysms that had ruptured intraoperatively, and the increased morbidity and mortality associated with this event were detailed as well. In each case, the family was given the option of declining the use of rFVIIa. In this unique subgroup of patients, we felt the use of the drug was justified on a clinical basis given the poor condition of the patients and the fact that a significant intraoperative rupture would likely be devastating in this setting.

Results

Pertinent clinical information including patient age, sex, Hunt-Hess grade, aneurysm locations, complications, and patient outcomes at 6 months is summarized in Table 1. In this series, 13 patients had suffered multiple hemorrhages prior to surgery. Four patients had large and complex lesions that were expected to require extensive microdissection prior to adequate clip placement. One patient had a presumed neck bleed based on aneurysm morphology, and it was felt that it would be difficult to achieve adequate proximal control prior to isolation of the aneurysm neck in 2 cases.
Table 1

Pertinent clinical information and outcomes in 18 patients with ruptured intracranial aneurysms who received perioperative rFVIIa

Patient

Age

Sex

H/H Grade

Aneurysm location(s)

Complication

VP shunt

Vasospasm

Outcome at 6 months

1

68

F

V

PCoA

Pneumonia

Yes

Yes

Death, 3 months

2

60

F

IV

MCA

Pneumonia, sepsis

No

Yes

Death, 3 weeks

3

50

M

II

ACoA

Pneumonia, PICCO

Yes

No

Minor Disability

4

41

F

V

ACoA

UTI, PICCO

No

Yes

Good

5

85

F

II

MCA

Pneumonia

No

No

Good

6

71

M

V

ACOA

Pneumonia, sepsis, PICCO

No

Yes

Death, 1 month

7

55

F

IV

PCoA

PICCO

No

Yes

Minor Disability

8

82

F

IV

MCA, ICA Bif

Pneumonia

Yes

No

Moderate Disability

9

49

F

III

ACoA, DACA

Yes

Yes

Good

10

50

M

II

ACoA

Yes

No

Good

11

42

M

V

MCA, ACoA

DVT, PICCO

Yes

No

Moderate Disability

12

59

M

IV

MCA

PICCO

No

No

Good

13

57

F

II

ACoA

No

No

Minor Disability

14

72

M

III

ACoA

No

No

Death, 2 months

15

43

F

II

ICA Bif

PICCO

No

No

Good

16

60

F

II

PCoA, COA

No

No

Good

17

30

M

II

ICA Bif

No

No

Good

18

49

F

II

PCoA, MCA, DACA

UTI

Yes

No

Good

H/H grade = hunt/hess grade, PCoA = posterior communicating artery, MCA = middle cerebral artery, ACoA = anterior communicating artery, ICA Bif = internal carotid artery bifurcation, DACA = distal anterior cerebral artery, PICCO = peripherally inserted central catheter occlusion, DVT = deep venous thrombosis, UTI = urinary tract infection

Based on weight, the majority of patients (62.5%) were treated with the higher 4.8 mg dose of rFVIIa. Because of the bias in this series toward patients who had suffered multiple hemorrhages prior to surgery, this group included a high percentage (50%) of Hunt/Hess Grades IV and V hemorrhages (Table 1). Fifteen of 18 patients presented with acute hydrocephalus or evidence of raised ICP and underwent ventriculostomy placement. This high percentage of patients requiring early CSF diversion is also reflected in the high number of patients eventually requiring permanent ventriculoperitoneal shunting (Table 1). The diagnosis of hydrocephalus in this series was based on the finding of ventriculomegaly combined with raised intracranial pressure at the time of external ventricular catheter placement. The need for ultimate conversion to a permanent shunt was related to higher grade hemorrhage on presentation with more blood breakdown products.

One lower extremity deep venous thrombosis and seven upper extremity venous thromboses in association with peripherally inserted central catheter (PICC) lines were identified on screening Doppler evaluations. There were no associated pulmonary emboli. The incidence of venous thrombosis was higher due to higher surveillance in our Neurocritical care unit (NIC), with ultrasound screening performed routinely on patients 48–72 h after admissions. The majority of the thromboses were related to peripheral inserted central lines (PICC) and were at the line site. Our NIC quality control prior to this review did show a higher incidence of upper extremity DVTs with PICC insertion.

All patients underwent microsurgical aneurysm clipping via a standard pterional craniotomy by a single neurosurgeon (ESN). An orbitozygomatic approach was utilized in two patients with complicated anterior communicating artery aneurysms. In all cases, the surgeon noted that bleeding was well controlled throughout the operation. No patient suffered an intraoperative rupture, and in several cases, the surgeon was impressed subjectively by the amount of dissection tolerated by the recently ruptured lesions.

Discussion

In large series, intraoperative aneurysm rupture has been reported in up to 34% of cases and can significantly increase the morbidity and mortality rates in patients undergoing microsurgical aneurysm clipping procedures [7, 17, 19, 38]. Attempts to stop the bleeding from an intraoperative rupture without the advantage of clear visibility can result in serious injury to local neurovascular structures. In the setting of an intraoperative rupture, gentle tamponade with induced hypotension may control the bleeding in some cases, and temporary arterial occlusion represents an excellent option for addressing more significant ruptures [3, 9, 17, 30, 38]. We have previously described the use of adenosine to induce transient cardiac standstill in the face of a severe intraoperative rupture when the above techniques were unsuccessful [30]. Nevertheless, most surgeons would agree that it is highly preferable, for the sake of both the patient and surgeon, to avoid an intraoperative rupture entirely. Any measure that might decrease the risk of such intraoperative bleeding is therefore of great interest to microvascular neurosurgeons, and the potential use of rFVIIa for this purpose represented the basis of our study.

Antifibrinolytics in Subarachnoid Hemorrhage

The idea of utilizing antifibrinolytic agents in the setting of acute SAH is not new. In the early 1970s, there was a substantial amount of interest in the use of antifibrinolytic agents to prevent early aneurysm rebleeding following SAH [10, 2729, 41]. At that time, most centers were delaying primary aneurysm treatment, and therefore rebleeding represented a substantial source of morbidity and mortality in patients who were often waiting several weeks before undergoing surgical aneurysm repair. Multiple early studies demonstrated the ability of antifibrinolytic agents such as epsilon-aminocaproic acid and tranexemic acid to reduce the incidence of early recurrent hemorrhage following SAH [10, 41]. Unfortunately, subsequent studies showed significant increases in the rates of vasospasm, hydrocephalus, and pulmonary thromboemboli in patients receiving antifibrinolytic agents that generally offset the benefit of the early decreased rebleeding rates [1, 42, 44].

More recently, various investigators have examined the safety and efficacy of shorter courses of antifibrinolytic therapy combined with early aneurysm treatment to decrease aneurysm rebleeding rates [13, 18, 36, 37, 46]. It has been suggested that briefer courses of antifibrinolytic therapy in association with early aneurysm treatment could minimize the risk of rebleeding while limiting the potential adverse effects of the antifibrinolytic agents. A recent comprehensive review by Chwajol et al. [5] concluded that short-term use of antifibrinolytic agents (less than 3 days) in the presence of calcium channel blocking agents is a reasonable strategy to prevent acute rebleeding and improve long-term outcome in patients with acute SAH. To our knowledge, no study has evaluated the use of an antifibrinolytic agent administered specifically in an attempt to decrease the rate of intraoperative aneurysm rupture.

Recombinant Activated Factor VII and Neurosurgery

Recombinant activated Factor VII has been approved for use by the Food and Drug Administration in the setting of Hemophilia A or B with acquired inhibitors to factor VII and IX, in patients with factor VII deficiency, or in those with Glanzmann thrombasthenia [11, 12, 24, 31]. Other indications include coagulopathy related to defective platelets or thrombocytopenia [2, 6, 8, 24]. rFVIIa works at the site of vascular injury, binding to activated platelets adherent to the wall of the damaged vessel, and thereby generating thrombin and initiating coagulation [14, 32]. The drug half-life is only 2.3 h, and onset of effect is observed within 20 min of administration [6, 22]. Although the potential usefulness of rFVIIa has gained increasing recognition in the neurosurgical literature [8, 11, 33], enthusiasm for its use has been tempered by safety concerns including the potential for thromboembolic complications [34, 40]. Any study evaluating the use of rFVIIa in SAH will have to address the possibility of an increased risk for cerebral vasospasm, hydrocephalus, or thromboembolic sequelae.

In 2000, Ingerslev [15] described the use of rFVIIa to treat a patient with hemophilia requiring craniotomy for an acute epidural hematoma. Subsequently, Veschev et al. [45] administered rFVIIa to a patient with an acute subdural hematoma and an international normalized ratio (INR) of 6.39. To rapidly correct the coagulopathy, a single dose of 120 µg/kg of rFVIIa was used, and the drug was found to be safe and effective. Lin et al. [20] reported the successful use of rFVIIa in four patients with warfarin induced coagulopathy and either intraspinal or subdural hematoma. Acceptable hemostasis was achieved rapidly after administration of 16 to 22 µg/kg of rFVIIa, and no thrombotic complications were encountered. Similar experiences were subsequently reported demonstrating the effectiveness of the drug without associated complications [26, 33]. Further work by Sorensen et al. [39] suggested that smaller doses, ranging from 10 to 40 µg/kg, could be equally effective while carrying lower theoretical risks of thrombotic complications.

In a more recent study evaluating the use of rFVIIa to prevent aneurysmal rebleeding, Pickard et al. [34] enrolled 10 patients in an open label, dose escalation safety trial. Two patients received an 80 µg/kg single bolus, two received an 80 µg/kg bolus followed by continuous infusion at 3.5 µg/kg/h, and one patient received an infusion of 7 µg/kg. Five patients received a control infusion. Unfortunately, one patient developed middle cerebral artery thrombosis contralateral to the treated aneurysm, and the study was suspended. This complication, if related to the rFVIIa, may have resulted from the high doses utilized, the long duration of dosing, and the use of repeated boluses. As noted previously, smaller doses may be equally beneficial, and a single perioperative dose has the benefit of a short half-life with rapid onset of action and limited potential for systemic complications [22, 39]. Since the time of the suspension of Pickard’s study, no new systematic evaluation of the use of rFVIIa in the setting of SAH has been attempted.

Limitations of the Present Study

We hypothesized that a single dose of rFVIIa might be beneficial in terms of decreasing the risk of intraoperative rupture without carrying the increased risks of the serious side effects that have been seen with the more prolonged use of antifibrinolytic agents. We have not used rFVIIa at the time of admission in all SAH patients to prevent early rebleeding because many of our patients will undergo endovascular aneurysm treatment, and the powerful thrombotic action of this agent may have unpredictable effects on the anticoagulation and antiplatelet therapy utilized during coiling procedures.

This study is only a preliminary safety evaluation of the use of rFVIIa in the setting of SAH. The small size of the study population will obviously limit the statistical meaningfulness of our results. In addition, because the study population included a high percentage of poor grade SAH patients, the incidences of vasospasm (33%), and hydrocephalus requiring VP shunting (39%) are in keeping with reported rates for this population [16, 21, 43]. As expected, in our experience the incidence of these problems was higher in those patients with more severe hemorrhages. Nevertheless, it is unclear whether these complications would still occur more frequently than expected in a larger cohort of “good grade” SAH patients treated with rFVIIa. To better address this question, we would suggest that a larger trial may be appropriate. In particular, a randomized, multicenter trial evaluating the use of rFVIIa in an unselected cohort of SAH patients undergoing microsurgical aneurysm clipping would be a critical step in better defining the complications associated with and the potential role of rFVIIa in this setting.

Conclusion

We reviewed the records of 18 patients with SAH who received rFVIIa in an attempt to diminish the incidence of intraoperative aneurysm rupture in a group felt to be at high risk for this potentially severe complication. In this series, the use of rFVIIa was safe, and we encountered no intraoperative ruptures. Based on these results and a review of the existing literature, a larger controlled trial may be warranted to confirm these findings and to properly test efficacy in a larger population.

Copyright information

© Humana Press Inc. 2008