Effects of Tisseel fibrin glue on the central nervous system of nonhuman primates.

Abstract

For many years, neurosurgeons and otolaryngologic surgeons have used the fibrin glue product Tisseel to repair skull-base spinal fluid leaks and to help secure repairs following anterior cranial-base surgery. Despite the widespread use, the potential focal cerebral toxicity of this fibrin glue has never been investigated. We studied the safety of Tisseel applied directly to neural tissue (brain parenchyma, cervical cord, and C3-C6 spinal roots) of 6 monkeys (Macaca nemestrina) to determine if any underlying biochemical injury would occur. Another 3 animals that served as controls received saline rather than Tisseel. We found that median nerve electroencephalographic tracings and somatosensory evoked potentials in the experimental and control animals were identical. Likewise, cerebrospinal fluid indicators of neuronal or brain injury, inflammatory responses, and infection were negative in both groups. Finally, there were no significant differences between the two groups with respect to edema volumes and apparent diffusion coefficient values. We conclude that Tisseel does not induce an apparent inflammatory response or abnormal neurophysiologic or histologic response within 5 days of its application when it is applied directly to the brain parenchyma or onto the cervical spinal cord.

Introduction

The demonstrated effectiveness and safety of fibrin glue (thrombin-containing material) in sealing dural leaks has led to its widespread use in clinical neurosurgical procedures. (1-11) Tisseel (a fibrin sealant) was approved by the United States Food and Drug Administration in 1988, (12) despite the fact that studies in animals exposed to thrombin revealed that thrombin causes brain edema, (13,14) neuronal injury, (15) apoptosis, (15) and seizures. (16) In addition, inhibition of factor V and bovine thrombin, which can develop following thrombin exposure, reportedly induced coagulopathy and inflammatory reactions, (17-21) Although these studies were not conducted with Tisseel, they raised the specter that Tisseel might cause similar adverse effects in view of its human thrombin constituent. (12,22)

It is important that Tisseel's effects be studied in nonhuman primates because their blood groups (23,24) and coagulation cascade (25,26) are similar to those of man. Use of the nonhuman primate-sized brain also allows for higher resolution on magnetic resonance imaging (MRI). In this article, we describe our study of the effects of Tisseel applied directly to the brain parenchyma, the cervical cord, and spinal nerves of 6 monkeys.

Materials and methods

Animals. Following a protocol approved by the Institutional Animal Care and Use Committee at our institution (protocol #0010896), we quarantined 9 research-naive, pigtail monkeys (Macaca nemestrina) of both sexes for 30 days. The animals weighed between 5 and 7 kg. They were free of tuberculosis, hepatitis, and simian immunovirus and retrovirus. The monkeys were maintained in a temperature-controlled environment (72 to 75[degrees]F) with a 12-hour light/12-hour dark cycle. They were fasted overnight prior to the study, but they had free access to water. Six monkeys were included in the Tisseel group, and the remaining 3 served as controls.

Anesthesia and surgical procedures. All 9 monkeys were anesthetized with 10 mg/kg of ketamine and 1 mg of atropine intramuscularly. A peripheral intravenous catheter was inserted percutaneously into the saphenous vein, and the monkeys were intubated with a cuffed endotracheal tube. One gram of IV cefazolin was administered as prophylaxis against infection. The monkeys were immobilized with 0.06 mg/kg of IV pancuronium bromide and anesthetized with 0.6 to 1.0% isoflurane, 66% nitrous oxide, and 33% oxygen. They were mechanically ventilated during surgery. Using an aseptic technique, we inserted a central venous catheter into the femoral vein for drug infusion, and we placed a femoral arterial line to monitor arterial blood pressure and to obtain arterial blood samples (0.5 ml) for pH and blood gas measurement.

The dura was incised via a left frontal craniotomy 2 cm in diameter. Corticectomy (1 cm in length and depth) was performed in the middle frontal gyrus for direct subpial placement of the Tisseel into the brain parenchyma. In the 6 experimental monkeys, the corticectomy was filled with Tisseel glue (~1.0 ml), the dural defect was sealed with Tisseel, and the cranial defect was sealed with bone wax. In each control monkey, the corticectomy was filled with saline and the cranial defect with bone wax. The galeoperiosteal flaps were sutured in place, and the scalp was stapled closed.

Laminectomies at C3-C4 were performed on 7 monkeys--4 in the Tisscel group and the 3 controls. Tisseel (2 ml) was injected into the right lateral intrathecal space through a 25-gauge intracath, which extended the passage of Tisseel down to the C6 level. The dural defect was sealed with Tisseel in the 4 experimental animals and left unsealed in the 3 controls.

Electroencephalography and somatosensory evoked potentials. Median nerve electroencephalographic (EEG) tracings and somatosensory evoked potentials (SSEPs) were recorded prior to surgical intervention under ketamine anesthesia, then continuously for the first 24 hours following the application of Tisseel. The goal of monitoring was to evaluate spinal cord, cervical nerve root, and brain function for evidence of seizures. During the 24-hour monitoring period, the isoflurane/nitrous oxide anesthesia was replaced by anesthesia with 25 [micro]g/kg/hr of IV fentanyl; if an animal could not maintain a systolic arterial blood pressure of less than 140 mm Hg under fentanyl, a 2.5 mg/hr bolus of diazepam was administered. At 3 and 5 days postoperatively, repeat EEG and SSEP recordings were obtained under ketamine anesthesia.

Four-channel bihemispheric EEG recordings were obtained via subperiosteal needle electrodes. SSEP recordings were obtained via bilateral median nerve stimulation performed independently on each limb. Scalp electrodes were placed at P4/Fz and P3/Fz (according to a modified international 10-20 system), and the subcortical electrode was placed at the mastoid and referenced to Fz. SSEPs were elicited by producing sufficient intensity to evoke a consistent response at a stimulation frequency of 3.43 Hz and a duration of 0.2 msec. Band-pass filters were set between 3 and 300 Hz with a gain of 20,000 for cortical recordings and between 30 Hz and 1 kHz with a gain of 50,000 for cervical recordings. Averages were computed for 128 trials, and at least two averages were computed for each limb. Averages for the controls were obtained to verify the lack of contamination by nonbiological noise. EEG and SSEP data were reviewed by a neurophysiologist (J.B.) who had been blinded to each primate's treatment.

Twenty-four-hour monitoring and 5-day observation. In addition to EEG and SSEP, we also continuously monitored blood pressure, rectal temperature, cardiac rhythm by electrocardiography, and arterial oxygenation by pulse oximetry. To ensure normal values, arterial blood gases were measured prior to extubation and prior to transport of each subject to the recovery room. Throughout the 5-day observation period, the monkeys were kept in cages while veterinary technicians provided 24-hour coverage to detect behavioral signs of seizures.

MRI evaluation. MRIs of the brain and cervical spine were performed to obtain apparent diffusion coefficient (ADC) values on diffusion-weighted imaging (DWI) and T2-weighted imaging. Scans were obtained before surgery and 5 days later to assess edema (extracellular and intracellular water). All MRI scanning was performed with the monkeys in the supine position inside a 3.0 Tesla whole-body imager (General Electric Medical Systems; Milwaukee) operating under LX 8.3 software and equipped with broad-band and echo-planar imaging (EPI) capabilities.

Imaging was obtained by sagittal T1-weighted spinecho and axial proton-density and T2-weighted fast-spinecho sequences supplemented by axial fluid-attenuated inversion recovery (FLAIR) EPI and T2-weighted gradient-echo sequences. Although the anatomic resolution and soft-tissue contrast is exceptional at 3.0 Tesla, acute ischemic changes cannot be detected. Therefore, we used DWI.

ADC mapping was obtained by diffusion-weighted, ccrebrospinal fluid-(CSF) nulled, inversion-recovery EPI sequence optimized for isotropic diffusion weighting. Diffusion weighting was incorporated in the standard fashion by applying pulsed gradients (duration: 50 msec) symmetrically about the 180[degrees] refocusing pulse alternatively in the x, y, and z directions. ADC trace maps were calculated by voxel-wisc logarithmic fitting of the directional mean signal intensity as a function of b-value.

Regional measurements of diffusion weighting, ADC, and edema volume calculated from the T2-weighted images were analyzed as a function of time after Tisseel placement on the brain and spinal cord to determine whether the Tisseel had any effects on tissue fluid content, tissue permeability, and tissue perfusion adjacent to or remote from the locations of its application.

CSF analyses by enzyme-linked immunosorbent assay (ELISA). CSF samples (3 ml) were obtained by lumbar puncture before surgery and by cisterna magna puncture on postcorticectomy day 5 prior to perfusinn fixation of the brain. CSF samples were analyzed by ELISA for the following values:

* interleukin-6 (IL-6) (R&D Systems; Minneapolis), an indicator of the inflammatory response to the fibrin glue

* neuron-specific enolase (NSE) (Syn-X Pharma; Mississauga, Ont.), an indicator of neuronal injury

* S-100B protein (Skye PharmaTech; Mississauga), an indicator of brain injury

Histologic analysis. On postcorticectomy day 5, the monkeys were anesthetized with 40 mg/kg of IV sodium pentobarbital, and the descending aorta was cross-clamped via a midsternal thoracotomy. A 16-gauge catheter was inserted into the left cardiac ventricle for perfusion with 500 ml of heparinized saline (10 IU/ml) and 500 ml of 2% paraformaldehyde. The brain and the cervical spinal cord (C1-T1) were removed and placed in ajar containing 10% buffered formalin (Fischer Scientific Co.; Fair Lawn, N.J.).

Hematoxylin and eosin-stained sections were analyzed for eight histopathologic changes: (1) the presence or absence of glue at the lesion site, (2) leptomeningeal/ cortical hemorrhage, (3) the presence or absence of neutrophils in the leptomeninges and at the lesion site, (4) the pattern of edema, (5) capillary proliferation and distribution, (6) astroglial reaction, (7) the presence of macrophages, and (8) ischemic neuronal change or necrosis in the adjacent cortex.

Statistical analysis. The Prophet software (AbTech Corp.; Charlottesville, Va.) was used for statistical analysis by one-way analysis of variance (unblocked) and by Tukey multiple comparison and Student's t tests for comparisons (statistical significance: p<0.05).

Results

Twenty-four-hour intensive monitoring. Throughout the study, arterial blood pressures and rectal temperatures were maintained within normal limits. All physiologic variables were unremarkable, and the monkeys tolerated the procedures well. There was no evidence of epileptiform activity on EEG throughout the 24 hours. Diazepam was needed only sparingly; of the 9 monkeys, 5 did not require any diazepam (3 Tisseel subjects and 2 controls). When diazepam was administered, 2 Tisseel monkeys received 2 doses of 2.5 mg, 1 Tisseel monkey received 1 dose of 2.5 mg, and l control received 2.0 mg.

Five-day postsurgical observations. None of the monkeys exhibited behavioral signs of seizures. SSEP readings were unchanged in all monkeys, and there were no significant differences between pre- and postoperative recordings and no differences between the two groups (table 1). The only abnormality observed was a slight monoparesis of the right upper extremity of one of the Tisseel-treated monkeys; this abnormality was probably secondary to nerve root trauma during intrathecal Tisseel injection, as suggested by a twitch of the monkey's arm during the injection.

CSF analyses. CSF anaerobic and aerobic cultures (table 2, top) and Gram's stains were negative for bacteria and white cells in all samples obtained prior to and 5 days after Tisseel application, indicating that there was no inflammatory response or infection. CSF glucose and protein values prior to and 5 days following corticectomy and laminectomy with and without Tisseel application were unremarkable. In two of the samples obtained from monkeys in the fibrin group, elevated protein levels were clearly attributable to traumatic CSF sampling, as indicated by bloody CSF.

Findings on ELISA analyses of CSF for IL-6, NSE, and S-100B protein were not significantly different after the application of Tisseel in the experimental monkeys nor in the controls than they were beforehand (table 2, bottom).

Edema volume by T2-weighted MRI. Edema volume was assessed by density thresholding of the T2-weighted images (figure 1) in the Tisseel and control monkeys (table 3). No significant difference in edema volume between the two groups was observed.

[FIGURE 1 OMITTED]

ADC by DWI. DWI showed that there were no significant differences in ADC values between the edematous regions treated with Tisseel and those that were not treated.

Histopathology. The histologic changes observed in the regions of the corticectomy with Tisseel could be seen in the cortical mantle and underlying subcortical white matter (figure 2). The area surrounding the corticectomy with Tisseel was characterized by edema, capillary infiltration, and a lack of neutrophils (figure 3); the lack of neutrophils is important because it indicates the absence of an inflammatory response.

[FIGURES 2-3 OMITTED]

Control corticectomies exhibited a similar degree of tissue necrosis and perhaps slightly more edema than that observed with Tisseel (figure 4, A). The degree of capillary proliferation and neuronal necrosis was the same (figure 4, B).

[FIGURE 4 OMITTED]

Axial sections of the C6 spinal cord show that the Tisscel injected intrathecally at C3-C4 extended down to C6 (figure 5, A). No tissue reaction was evident alter histologic examination of the spinal cord in the regions where Tisseel was applied (figure 5, B).

[FIGURE 5 OMITTED]

Discussion

Fibrin sealants are commonly used by neurosurgeons, primarily to repair dural defects, augment dural repairs, and achieve hemostasis, both intracranially and in the spinal canal. Despite this widespread use and no reports of adverse effects from the application of fibrin glue directly to the CNS, (4,27) the potential for neurotoxicity has not been previously studied in nonhuman primates. (13,14,16)

Thrombin, a serine protease enzyme, is activated from prothrombin by factors Xa and Va. (19,28) In its primary role, thrombin converts fibrinogen to fibrin monomer, which polymerizes to form fibrin matrix. Thrombin triggers a host of other effects on cells through protease-activated receptors, which use G proteins to activate various cellular responses (28) that promote coagulation, clot formation, changes in endothelial cell shape, and endothelial permeability. These responses, in turn, promote local transudation of proteins and fluid, activation of local inflammatory responses, and generation of reactive oxygen species. Based on these effects, thrombin applied directly to or into the brain might be expected to induce a variety of pathologic responses, including edema, seizures, and apoptotic changes. (13-16) However, these effects have not been observed following the use of Tisseel in humans--and, as a result of our study, they have now been demonstrated to be absent in nonhuman primates as well. The reason for the lack of an adverse effect from Tisseel may be that the act of premixing thrombin and aprotinin (component 2) with fibrinogen and factor XIII (component 1) (29) mitigates a direct thrombin effect on the tissue. (30)

We found no evidence of clinical seizure activity following the application of Tisseel directly into the brain parenchyma. This finding differs from results obtained by Lee et al, who injected thrombin into rat basal ganglia. (13) A possible explanation for the difference between their findings and ours is that Lee et al injected pure bovine thrombin, whereas we applied Tisseel after premixing the thrombin and aprotinin with fibrinogen and factor XIII. It is unclear whether or not this difference could be responsible for our contrasting findings, but the 24 hours of continuous EEG recording postapplication should have allowed us sufficient time to detect any seizures directly attributable to Tisseel, even in the face of the occasional administration of low-dose diazepam to some of our study animals.

The fact that we used diazepam injected as bolus doses of 2.0 to 2.5 mg once or twice in 4 of the 9 monkeys might have masked evidence of possible seizure activity. Diazepam was used in some circumstances to keep systolic arterial blood pressure below 140 mm Hg without a need for excessive doses of fentanyl, which has been implicated in the generation of seizures and brain damage in rats. (31-33) However, in our study, 3 of the 6 monkeys in the Tisseel group did not receive diazepam, while 2 received only 2 doses of 2.5 mg and 1 received 1 dose of 2.5 mg over the 24 hours. None of these monkeys exhibited any seizure or epileptiform activity.

Perhaps more problematic than Tisseel's potential effects on epileptiform activity, edema, or apoptotic changes in the CNS is its potential to induce antibody formation against thrombin of either bovine or human origin. (17,18,20,21) As both an enzyme and a protein, thrombin could induce an immune response by generating antibodies to itself and to other components of the coagulation cascade (e.g., fibrinogen and factor V), which might then cause coagulopathies. Because our monkeys experienced only a single exposure to Tisseel, our study does not settle this issue. Other studies of the effects of thrombin in inducing edema, isehemia, and antibody formation involved bovine thrombin rather than Tisseel. Studies of repeated Tisseel exposures would be needed to clarify this issue.

Our analysis of CSF for IL-6, NSE, and S- 100B protein levels was conducted to look for an inflammatory response and neuronal damage:

* An increase in IL-6 level is an indicator of an inflammatory response. (34-38) Our 5-day observation period was sufficiently long to enable us to observe any increases in IL-6 had there been any.

* NSE--a 78-kDa protein that originates primarily in neurons and neuroendocrine cells--should be a sensitive marker for neuronal degradation. (21) Increased CSF levels of NSE have been found in dogs and rats subjected to cardiac arrest, (21,29) Creutzfeldt-Jakob disease, (39) head injury, (40,41) and focal ischemia. (15)

* S-100B protein--an acidic calcium-binding protein that is present in high concentrations in glial and Schwann's cells--is released into serum and CSF after CNS cell degradation. It has been shown to be an indicator of CNS damage after cardiac surgery. (29)

Neither IL-6, NSE, nor S-100B protein levels were elevated in the CSF samples we drew before and after corticectomy and application of Tisseel to the brain. These findings concur with our histopathologic observations. According to these findings, Tisseel does not appear to induce any more tissue inflammation, edema, or ischemic injury than does corticectomy alone.

In conclusion, the results of our study show that Tisseel causes no adverse effects when it is placed directly into the brain and onto the spinal cord of nonhuman primates. Based on our findings, we conclude that Tisseel is benign with respect to acute interactions with the human brain and spinal cord.

Table 1. Neuroohvsiologic data on Tisseel and control monkeys

                                         Preop

Tisseel monkeys (n = 6)  LAT (msec)      7.5 [+ or -] 0.3
  Right arm SSEP         AMP ([micro]V)  5.6 [+ or -] 2.3

Left arm SSEP            LAT (msec)      7.4 [+ or -] 0.3
                         AMP ([micro]V)  5.4 [+ or -] 0.7

Controls (n = 3)         LAT (msec)      7.5 [+ or -] 0.4
  Right arm SSEP         AMP ([micro]V)  3.3 [+ or -] 1.4

Left arm SSEP            LAT (msec)      7.6 [+ or -] 0.6
                         AMP ([micro]V)  2.7 [+ or -] 0.3

                                         Day 3

Tisseel monkeys (n = 6)  LAT (msec)      7.5 [+ or -] 0.3
  Right arm SSEP         AMP ([micro]V)  4.8 [+ or -] 1.4

Left arm SSEP            LAT (msec)      7.4 [+ or -] 0.5
                         AMP ([micro]V)  5.0 [+ or -] 0.3

Controls (n = 3)         LAT (msec)      7.4 [+ or -] 0.5
  Right arm SSEP         AMP ([micro]V)  5.0 [+ or -] 0.0

Left arm SSEP            LAT (msec)      7.5 [+ or -] 0.4
                         AMP ([micro]V)  2.5 [+ or -] 2.1

                                         Day 5

Tisseel monkeys (n = 6)  LAT (msec)      7.4 [+ or -] 0.2
  Right arm SSEP         AMP ([micro]V)  4.7 [+ or -] 0.5

Left arm SSEP            LAT (msec)      7.3 [+ or -] 0.3
                         AMP ([micro]V)  5.1 [+ or -] 1.0

Controls (n = 3)         LAT (msec)      7.5 [+ or -] 0.3
  Right arm SSEP         AMP ([micro]V)  3.5 [+ or -] 2.2

Left arm SSEP            LAT (msec)      7.7 [+ or -] 0.3
                         AMP ([micro]V)  2.8 [+ or -] 2.0

None of these differences is statistically significant (p<0.05).
Key: SSEP = somatasensory evoked potentials; LAT = latencv;
AMP = amplitude.

Table 2. Results of CSF analyses before and after
Tisseel or saline application

                            Preop
                 Glucose              Protein

Tisseel (n = 6)  66 [+ or -] 3.0      571 [+ or -] 746
Control (n = 3   69 [+ or -] 2.0      23 [+ or -] 9.0

                 IL-6                 NSE

Tisseel (n = 6)  10.7 [+ or -] 10.6   9.4 [+ or -] 6.9
Control (n = 3   7.6 [+ or -] 15      14.2 [+ or -] 14.6

                  Preop

                 Culture

Tisseel (n = 6)  Negative
Control (n = 3   Negative

                 S-100B

Tisseel (n = 6)  2.0 [+ or -] 1.0
Control (n = 3   2.4 [+ or -] 1.6

                             Day 5
                   Glucose            Protein

Tisseel (n = 6)  87 [+ or -] 45       122 [+ or -] 149
Control (n = 3   61 [+ or -] 15       59 [+ or -] 29

                 IL-6                 NSE

Tisseel (n = 6)  25.9 [+ or -] 4.8    26.2 [+ or -] 26.5
Control (n = 3   14.0 [+ or -] 23.4   20.1 [+ or -] 27.6

                  Day 5

                  Culture

Tisseel (n = 6)   Negative
Control (n = 3    Negative

                  S-100B

Tisseel (n = 6)   1.7 [+ or -] 1.6
Control (n = 3    2.0 [+ or -] 2.4

None of these differences is statistically significant (p<0.05).

Glucose concentrating are expressed as mg %, protein as mg/ml,
interleukin-6 (IL-6) as pg/ml, neuron-specific enolase (NSE)
as ng/ml, and S-100B

Table 3. ADC values in edematous brain regions surrounding
corticectomies in 4 Tisseel monkeys and 3 controls

           Preop ADC            Postop ADC
           ([10.sup.-5]         (10-5                5-day edema
            [cm.sup.2]/sec)     [cm.sub.2]/sec)      volume (ml)

Tisseel
(n = 4)                                              1.47 [+ or -] 0.80

Fibrin-1   130.8 [+ or -] 53.5  87.2 [+ or -] 2.4
Fibrin-2   89.0 [+ or -] 2.5    65.3 [+ or -] 71.9
Fibrin-3   86.5 [+ or -] 1.8    89.6 [+ or -] 6.9

Control    84.3 [+ or -] 6.2    94.6 [+ or -] 6.9
(n = 3)                                              0.80 [+ or -] 0.45

None of these differences is statisticaliy significant (p<0.05).

References

(1.) Cain JE, Jr., Dryer RF, Barton BR. Evaluation of dural closure techniques. Suture methods, fibrin adhesive sealant, and cyanoacrylate polymer. Spine 1988;13:720-5.

(2.) Crockard HA. The transoral approach to the base of the brain and the upper cervical cord. Ann R Coil Surg Engl 1985;67:321-5.

(3.) D'Arrigo C, Landolt AM. Fibrin sealing of mucoperichondrial flaps in endonasal-transsphenoidal pituitary surgery: Technical note. Neurosurgery 1994;35:529-31.

(4.) de Vries J, Menovsky T, van Gulik S, Wesseling P. Histological effects of fibrin glue on nervous tissue: A safety study in rats. Surg Neurol 2002;57:415-22.

(5.) Hadley MN, Martin NA, Spetzler RF, et al. Comparative transoral dural closure techniques: A canine model. Neurosurgery 1988; 22:392-7.

(6.) Jackson MR, Airing BM. Fibrin sealant in preclinical and clinical studies. Curr Opin Hematol 1999;6:415-19.

(7.) Knoringer P. Use of fibrin glue for sealing and prophylaxis of cranial and spinal CSF fistulae: Indications, technique and results. In: Schlag G, Redl H, eds. Fibrin Sealant in Operative Medicine. Ophthalmology-Neurosurgery. Vol. 2. Berlin; Heidelberg: Springer-Verlag, 1986:148-56.

(8.) Lenzi B, Cantini R, Parenti G. Importance of the use of Tissucol (Tisseel) in the reparation of the dura. In: Schlag G, Redl H, eds. Fibrin Sealant in Operative Medicine. Ophthalmology-Neurosurgery. Vol. 2. Berlin; Heidelberg: Springer-Verlag, 1986:145-7.

(9.) Pomeranz S, Constantini S, Umansky F. The use of fibrin sealant in cerebrospinal fluid leakage. Neuroehirurgia (Stuttg) 1991;34: 166-9.

(10.) Van Velthoven V, Clarici G, Auer LM. Fibrin tissue adhesive sealant for the prevention of CSF leakage following transsphenoidal microsurgery. Acta Neurochir (Wien) 1991; 109: 26-9.

(11.) Waldbaur H, Fahlbusch R. The use of fibrin sealant in the neurosurgical treatment of lesions of the base of the skull. In: Schlag G, Redl H, eds. Fibrin Sealant in Operative Medicine. Ophthalmology-Neurosurgery, Vol. 2. Berlin; Heidelberg: Springer-Verlag, 1986:176-80.

(12.) Siedentop KH, Park JJ, Shah AN, et al. Safety and efficacy of currently available fibrin tissue adhesives. Am J Otolaryngol 2001;22:230-5.

(13.) Lee KR, Betz AL. Keep RF, et al. Intracerebral infusion of thrombin as a cause of brain edema. J Neurosurg 1995;83:1045-50.

(14.) Lee KR, Colon GP, Betz AL, et al. Edema from intracerebral hemorrhage: The role of thrombin. J Neurosurg 1996;84:91-6,

(15.) Gong C, Boulis N, Qian J, et al. Intracerebral hemorrhage-induced neuronal death. Neurosurgery 2001 ;48:875-82.

(16.) Lee KR, Drury I, Vitarbo E, Hoff JT. Seizures induced by intracerebral injection of thrombin: A model of intracerebral hemorrhage. J Nenrosurg 1997;87:73-8.

(17.) Banoinger H, Hardegger T, Tobler A, et al. Fibrin glue in surgery: Frequent development of inhibitors of bovine thrombin and human factor V. Br J Haematol 1993;85:528-32.

(18.) Chouhan VD, De La Cadena RA, Nagaswami C, et al. Simultaneons occurrence of human antibodies directed against fibrinogen, thrombin, and factor V following exposure to bovine thrombin: Effects on blood coagulation, protein C activation, and platelet function. Thromb Haemost 1997;77:343-9.

(19.) Patterson C. Stouffer GA, Madamanchi N, Runge MS. New tricks for old dogs: Nonthrombotic effects of thrombin in vessel wall biology. Circ Res 2001 ;88:987-97.

(20.) Rapaport SI, Zivelin A, Minow RA, et al. Clinical significance of antibodies to bovine and human thrombin and factor V after surgical

use of bovine thrombin. Am J Clin Pathol 1992;97:84-91.

(21.) Spero JA. Bovine thrombin-induced inhibitor of factor V and bleeding risk in postoperative neurosurgical patients. Report of three cases. J Neurosurg 1993;78:817-20.

(22.) Shekarriz B, Stoller ML. The use of fibrin sealant in urology. J Urol 2002;167:1218-25.

(23.) Moor Jankowski J, Socha WW. Blood groups of macaques: A comparative study. J Med Primatol 1978;7:136-45.

(24.) Socha WW. Moor-Jankowski J, Ruffie J. Blood groups of primates: Present status, theoretical implications and practical implications: A review. J Med Primatol 1984;13:11-40.

(25.) Hawkey C. Coagulation. fibrinolysis and platelet function. In: Fiennes RN, ed. Pathology of Simian Primates. Part I. Basel; New York: Karger, 1972:318-30.

(26.) Stolte LA. Pregnancy in the rhesus monkey. In: Bourne GH, ed. The Rhesus Monkey. Vol II. New York: Academic Press, 1975: 171-230.

(27.) Ghulam Muhammad AK, Yoshimine T, Maruno M, et al. Topical application of fibrin adhesive in the rat brain: Effects on different cellular elements of the wound. Neurol Res 1997; 19:84-8.

(28.) Coughlin SR. Thrombin signalling and protease-activated receptors. Nature 2000;407:258-64.

(29.) Sellman M, Ivert T, Ronquist G, et al. Central nervous system damage during cardiac surgery assessed by 3 different biochemical markers in cerebrospinal fluid. Scand J Thorac Cardiovasc Surg 1992;26:39-45.

(30.) Jaquiss RD, Ghanayem NS, Zacharisen MC, et al. Safety of aprotinin use and re-use in pediatric cardiothoracic surgery. Circulation 2002;106(suppl 1):190-4.

(31.) Cervantes M, Antonio-Ocampn A, Ruelas R, et al. Effects of diazepam on fentanyl-induced epileptoid EEG activity and increase of multineuronal firing in limbic and mesencephalic brain structures. Arch Med Res 1996;27:495-502.

(32.) Kofke WA, Garman RH, Stiller RL, et al. Opioid neurotoxicity: Fentanyl dose-response effects in rats. Anesth Analg 1996;83: 1298-1306.

(33.) Sinz EH, Kofke WA, Garman RH. Phenytoin, midazolam, and naloxone protect against fentanyl induced brain damage in rats. Anesth Analg 2000;91:1443-9.

(34.) Becket KJ. Targeting the central nervous system inflammatory response in ischemic stroke. Curr Opin Neurol 2001; 14:349-53.

(35.) Block F, Peters M, Nolden-Koch M. Expression of IL-6 in the ischemic penumbra. Neuroreport 2000;11:963-7.

(36.) Hill JK, Gunion-Rinker L, Kulhanek D, et al. Temporal modulation of cytokine expression following local cerebral ischemia in mice. Brain Res 1999;820:45-54.

(37.) Qureshi AI, Suri MF, Ling GS, et al. Absence of early proinflammatory cytokine expression in experimental intracerebral hemorrhage. Neurosurgery 2001;49:416-20.

(38.) Stelmasiak Z, Koziol-Montewka M, Dobosz B, Rejdak K. IL-6 and slL-6R concentration in the eerebrospinal fluid and serum of MS patients. Med Sci Monit 2001;7:914-18.

(39.) Aksamit AJ, Jr., Preissner CM, Homburger HA. Quantitation of 14-3-3 and neuron-specific enolase proteins in CSF in Creutzfeldt-Jakob disease. Neurology 2001;57:728-30.

(40.) Herrmann M, Curio N, Jost S, et al. Release of biochemical markers of damage to neuronal and glial brain tissue is associated with short and long term neuropsychological outcome after traumatic brain injury. J Neurol Neurosurg Psychiatry 2001;70: 95-100.

(41.) Woertgen C, Rothoerl RD, Brawanski A. Time profile of neuron specific enolase serum levels after experimental brain injury in rat. Acta Neurochir suppl 2000;76:371-3.

From the Center for the Assessment of Surgical Technology and the Copeland Neurosurgical Laboratories, University of Pittsburgh Medical Center.

Reprint requests: Amin Kassam, MD, Department of Neurological Surgery, Presbyterian University Hospital, 200 Lothrop St., Suite B-400, Pittsburgh, PA 15213. Phone: [phone]; fax: [phone]; e-mail: [e-mail]

Dr. Kassam is a paid consultant for Baxter Corp., and this study was financed by an educational grant from Baxter Corp.