J Oral Maxillofac Surg

65:630-634, 2007
The Effects of Hemostatic Agents
on Peripheral Nerve Function:
An Experimental Study
Alper Alkan, DDS, PhD,* Samet I

nal, DDS,
Mehmet Yildirim, PhD, Burcu Bas , DDS, and
Erdal Agar, PhD
Purpose: In the practice of oral and maxillofacial surgery, hemostatic agents are sometimes placed in
close proximity to peripheral nerves. In the present study, we evaluated immediate and delayed effects
of 4 hemostatic agents (oxidized regenerated cellulose, 5% colloid silver-added gelatine sponge, bovine
collagen, bone wax) on peripheral nerve function.
Materials and Methods: A total of 25 rat sciatic nerves were prepared, and the amplitudes were
recorded with a physiological data acquisition system. Animals were randomly assigned to 5 groups:
control, oxidized regenerated cellulose, gelatine sponge, bone wax, and bovine collagen. The rst hour
records are dened as immediate effects of these hemostatic agents on nerve function. The animals were
then allowed to recover for 4 weeks. At the end of this period, the same surgical and recording
procedures were performed. These nal records are dened as delayed effects of hemostatic agents on
nerve function. According to nerve conduction velocity (NCV) and compound action potential (CAP)
values of the experimental groups, early and delayed effects of each hemostatic agent were statistically
compared with Bonferroni corrected test (P .05).
Results: Statistically, NCV was signicantly reduced, and the CAP was signicantly increased 1 hour
after surgery (P .05) in the group of oxidized regenerated cellulose. However, there were no signicant
differences after 4 weeks compared with the rst records. In the gelatine sponge group, CAP was
signicantly increased 4 weeks after the application. In the bovine collagen and bone wax groups, NCV
and CAP values (1 hour and 4 weeks after the application) were not statistically signicant compared
with initial control records.
Conclusion: The present study shows that bovine collagen is the most suitable hemostatic agent
applicable for peripheral nerves.
2007 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 65:630-634, 2007
In the practice of oral and maxillofacial surgery, he-
mostatic agents are sometimes placed in close prox-
imity to peripheral nerves. Especially after the enucle-
ation of large cystic lesions and tumors or tumor-like
lesions, peripheral nerves can be visualized. To pre-
vent intraoperative and postoperative bleeding, local
hemostatic agents are frequently used. Any alteration
in nerve function that occurs postoperatively is usu-
ally attributed to the trauma of the surgical procedure
rather than the hemostatic agents. Much research has
been done on the effectiveness of different local he-
mostatic agents to control hemorrhage. But the ef-
fects of these agents on the peripheral neural tissues
have not been studied, to our knowledge, except 1
study about oxidized regenerated cellulose.
1
Therefore, we investigated immediate and delayed
effects of 4 hemostatic agents on peripheral nerve func-
tion in the present study: oxidized regenerated cellu-
lose, gelatine sponge, bovine collagen, and bone wax.
Materials and Methods
ANIMALS
Twenty-ve female Wistar rats weighing 220 to
270 g were used for this study. The rats were housed
*Associate Professor, Department of Oral and Maxillofacial Sur-
gery, Faculty of Dentistry, Erciyes University, Kayseri, Turkey.
Research Assistant, Department of Oral and Maxillofacial Surgery,
Faculty of Dentistry, Ondokuz Mays University, Samsun, Turkey.
Research Assistant, Department of Physiology, Faculty of Medicine,
Ondokuz Mays University, Samsun, Turkey.
Research Assistant, Department of Oral and Maxillofacial Surgery,
Faculty of Dentistry, Ondokuz Mays University, Samsun, Turkey.
Professor, Department of Physiology, Faculty of Medicine, On-
dokuz Mays University, Samsun, Turkey.
Address correspondence and reprint requests to Dr I

nal: On-
dokuz Mays, U

niversitesi Dis Hekimligi Fakltesi, 55139 Kurupelit,

Samsun, Turkey; e-mail: sameti@omu.edu.tr
2007 American Association of Oral and Maxillofacial Surgeons
0278-2391/07/6504-0006$32.00/0
doi:10.1016/j.joms.2005.12.076
630
2 animals per cage in a room and were fed specic rat
chow and water ad libitum. All experiments were
performed at the physiology laboratory of Ondokuz
Mays University Faculty of Medicine. Animals were
randomly assigned to 5 groups: control, oxidized re-
generated cellulose (Surgicel; Ethicon, Neuchatel,
Switzerland), gelatine sponge (Gelatamp; Roeko, Lan-
genau, Germany), bovine collagen (Lyostypt; Braun,
Melsungen, Germany), and bone wax.
Oxidized regenerated cellulose is a sterile absorb-
able knitted fabric prepared by the controlled oxida-
tion of regenerated cellulose. The fabric is white with
a pale yellow cast and has a faint, caramel-like aroma.
It is strong and can be sutured or cut without fraying.
In addition to its local hemostatic properties, it is
bactericidal in vitro against a wide range of Gram-
positive and Gram-negative organisms including aer-
obes and anaerobes.
A gelatine sponge to which 5% colloidal silver is
added has a porous foam structure that absorbs its
own weight in blood several times over, promotes
thrombocyte aggregation due to the large surface, and
lls the wound cavity. The gelatine sponge remains in
the wound and is completely absorbed within 4
weeks. The addition of colloidal silver has an antimi-
crobial effect while being nontoxic.
Bovine collagen is a wet-stable, felt-like hemo-
static agent and is composed of native absorbable
collagen brils of bovine origin. The contact be-
tween collagen and blood leads to platelet aggrega-
tion. The collagen structure leads to additional sta-
bilization of the cloth. It is completely absorbed
within 3 weeks.
Bone wax, a mixture of beeswax with isopropyl
palmitate, is the oldest and least expensive of the
agents. It is available in small sections and is easily
applied to an exposed osseous surface. Hemostasis is
achieved by mechanical tamponade, with no direct
effect on the coagulation cascade.
2
OPERATIVE PROCEDURE
The animals were anesthetized with intraperitoneal
injection of ketamine (100 mg/kg body weight) and
placed prone on the table, with both fore and hind
paws xed with rubber bands. The lateral aspect of
the right thigh, hip, and ank was then routinely
prepared, including trimming off the hair and antisep-
sis (20% iodine solution). The sciatic nerve was ex-
posed through a posterolateral longitudinal straight
incision going down from the greater trochanter to
the lateral condyle of the femur, followed by blunt
dissection between the gluteus maximus and quadri-
ceps muscles. The entire length of the nerve was
made visible and its 3 main distal branches, the com-
mon peroneal, tibial, and sural nerves, were carefully
identied in the popliteal fossa. We exposed approx-
imately 1.5 cm of the sciatic nerve above its trifurca-
tion without injuring the epineurium (Fig 1).
ELECTROPHYSIOLOGY
Body temperature was monitored by a rectal probe
and maintained at 37C with a homeothermic blanket
system (Harvard, Holliston, MA). The recordings of
nerve potentials were carried out using a physiologi-
cal data acquisition system (PowerLab 4/SP; ADInstru-
ments, Castle Hill, NSN, Australia). Bipolar hook elec-
trodes were used to stimulate the left sciatic nerve
and to record the nerve potentials. For the rst stim-
ulating point, a hook electrode was placed proximally
to the sciatic notch. The recording electrode was
placed distally to the sciatic nerve. After stimulating at
the rst point, the electrode was moved to the second
stimulating point, 1 cm distal to the rst point. Supra-
maximal stimuli (10 V) were delivered to the sciatic
nerve by 0.2 ms in duration, from a stimulator on a
PowerLab 4/SP data acquisition system. The responses
were amplied with an amplier (BIO Amp, ADInstru-
ments) and stored on a computer. Scope software (AD-
Instruments) was used for data capture and analysis.
The latencies were measured from the stimulus arti-
fact to the onset of the negative wave deection.
Nerve conduction velocities (NCV) were calculated
by dividing the distance between stimulating elec-
trodes by the average latency difference between the
onsets of the compound action potential (CAP). The
amplitude of the CAP was measured from peak to
peak. To determine the latency, electrical stimulation
was repeated 10 times and averaged per rat. A total
of 25 sciatic nerves were prepared and the ampli-
tudes were recorded as described above. Responses
were recorded after the identication of the nerve
in the popliteal fossa. Then the nerve was sur-
FIGURE 1. Identication of the sciatic nerve in the popliteal fossa.
Alkan et al. The Effects of Hemostatic Agents on Peripheral Nerve
Function. J Oral Maxillofac Surg 2007.
ALKAN ET AL 631
rounded by hemostatic agents of similar size (Fig 2).
The rst hour records were taken, and the wounds
were closed with 3/0 silk sutures. Before these re-
cordings, the nerve was irrigated with 0.9% NaCl
solution once every 10 minutes for the protection of
the tissue. The rst hour records were dened as
immediate effects of these hemostatic agents on
nerve function. The animals were then allowed to
recover for 4 weeks. At the end of this period, the
same surgical procedures were performed, and the
records were taken as described before. These nal
records were dened as delayed effects of hemostatic
agents on nerve function.
STATISTICAL ANALYSIS
Data were expressed as the mean SEM. For com-
parisons to the NCV and CAP values of the experi-
mental groups, 1-way analysis of variance (ANOVA)
was performed to evaluate statistical difference. Where
the overall signicance at P .05 was attained by
ANOVA, comparisons between individual group means
were made using the Bonferroni corrected test as
indicated (Figs 3, 4).
Results
In the bovine collagen and bone wax groups, NCV
and CAP values (1 hour and 4 weeks after the appli-
cation on the sciatic nerve) were not statistically sig-
nicant compared with initial control records.
In the gelatine sponge group, CAP was increased
statistically signicantly 4 weeks after the rst sur-
gery.
Statistically, NCV was signicantly reduced and the
CAP was signicantly increased 1 hour after the sur-
gery (P .05) in the oxidized regenerated cellulose
group. However, there were no signicant differ-
ences after 4 weeks compared with rst records.
Interestingly, partial necrosis and walking distur-
bances were seen on the operated legs of the rats
after 1 to 3 weeks in this group.
Discussion
Due to rich blood supply, severe hemorrhages can
be seen during the surgery in the maxillofacial region.
In this region, the vessels usually take place in the
FIGURE 2. Application of the hemostatic agents to the nerve.
A, Oxidized regenerated cellulose. B, Bone wax. C, Bovine collagen.
D, Gelatine sponge.
Alkan et al. The Effects of Hemostatic Agents on Peripheral Nerve
Function. J Oral Maxillofac Surg 2007.
FIGURE 3. Nerve conduction
velocity of different hemostatic
agent groups recorded from the
sciatic nerve at various intervals.
*Statistically signicant difference
(P .05).
Alkan et al. The Effects of He-
mostatic Agents on Peripheral
Nerve Function. J Oral Maxillo-
fac Surg 2007.
632 THE EFFECTS OF HEMOSTATIC AGENTS ON PERIPHERAL NERVE FUNCTION
associated neurovascular bundle (infraorbital, inferior
alveolar, mental, palatina descendens, incisivus, etc).
Most of the surgeons use suitable hemostatic agents
for hemostasis without knowing the effect of these
agents on nerve function. Patients whose neuro-vas-
cular bundle is visualized commonly experience tem-
porary or persistent sensorial disturbances in the
trace of these nerves after the operation. This is usu-
ally attributed to the result of the surgical trauma, but
the effect of these hemostatic agents on sensory dis-
turbances is still unknown. This study was performed
to investigate whether application of the hemostatic
agents is appropriate on peripheral nerves.
There is some information in the literature about
the effects of oxidized regenerated cellulose on pe-
ripheral nerve function. In addition to maxillofacial
surgery practice, it is also used effectively in neuro-
surgery, thoracic surgery, and orthopedics.
The electrophysiological results of the present
study conrm Brodbelt et als
3
suggestions because
they reported delayed paraplegia in all intraspinal and
perispinal procedures after the use of excess oxidized
regenerated cellulose. Thus, they stated that the over-
liberal use of it should be avoided, and attempts
should be made to remove all excess oxidized regen-
erated cellulose once adequate hemostasis is ob-
tained.
Iwabuchi et al
4
reported complete paraplegia
caused by oxidized regenerated cellulose 20 hours
after thoracotomy. Fifty days after the removal of the
oxidized regenerated cellulose, the patient began to
show improvement. The results of this study regard-
ing the temporary effect of the oxidized regenerated
cellulose are in accordance with that of a previous
report.
Several authors
5,6
have reported optic neuropathy
caused by oxidized regenerated cellulose. The possi-
ble mechanism of this nerve injury might be the
strong acidity and the nerve compression caused by
swollen oxidized regenerated cellulose.
To our knowledge, there is only 1 study about the
effect of oxidized regenerated cellulose on nerve
function in oral and maxillofacial surgery literature.
Loescher and Robinson
1
studied the effect of surgical
medications on peripheral nerve function to be used
in oral and maxillofacial procedures. They reported
variable effects of oxidized regenerated cellulose on
rat saphenous nerves. In their study, in 3 of 8 exper-
iments, nerve conduction was blocked within 30 min-
utes and no CAPs could subsequently be recorded. In
2 further experiments, no CAPs could be recorded at
75 and 105 minutes, respectively. In the remaining 3
experiments, although CAPs were recorded after 2
hours, their amplitudes were reduced when com-
pared with the initial recording.
1
On the contrary,
CAP was signicantly increased 1 hour after the ap-
plication of oxidized regenerated cellulose in the
present study. The difference between CAP values
might be explained by using different recording tech-
niques. They recorded NCV and CAP by using mois-
tened oxidized regenerated cellulose with Ringers
solution. The manufacturer (Ethicon) does not recom-
mend moistening the material with water or physio-
logical saline solution. This study indicates that oxi-
dized regenerated cellulose has an immediate effect
on peripheral nerves 1 hour after the surgery. The
FIGURE 4. Amplitude of the
compound action potentials of dif-
ferent hemostatic agent groups re-
corded from the sciatic nerve at
various intervals. *Statistically sig-
nicant difference (P .05).
Alkan et al. The Effects of He-
mostatic Agents on Peripheral
Nerve Function. J Oral Maxillo-
fac Surg 2007.
ALKAN ET AL 633
CAP and NCV regained characteristics that were sim-
ilar to those of controls after 4 weeks. This period was
2 weeks in Loescher and Robinsons experiment.
1
In the oxidized regenerated cellulose group of the
present study, partial necrosis was seen on the left
legs of all rats. Two possible scenarios might explain
this nding. In the rst scenario, the decrease in the
NCV may have resulted in numbness of the legs and
the rats may have bitten their legs. The second sce-
nario may be unattempted femoral artery cut or vas-
cular compression during surgery. To nd the answer
to these possibilities, we made another pilot study
and cut the femoral artery only after the exploration
of the sciatic nerve. Rats were still placed in the same
cage. No hemostatic agent was placed on the nerve in
this group. As a result, the partial necrosis was not
seen. Therefore, the peripheral necrosis might be
explained by the inhibitor immediate effect of oxi-
dized regenerated cellulose to the nerve. On the other
hand, further histomorphometric studies are needed
to explore this possibility.
There is no literature information about the effects
of gelatine sponge, bovine collagen, and bone wax on
nerve function. The present study shows that bovine
collagen and bone wax have more advantages than
oxidized regenerated cellulose and gelatine sponge
on peripheral nerve function. Delayed effects of gela-
tine sponge on nerve function were seen in this study.
CAP was signicantly increased 4 weeks after the
application in this group. The nerve function could be
affected by the addition of silver to gelatine sponge.
For this reason, further studies are needed to obtain
more reliable results.
Bone wax has a physical barrier effect on bone
trabecules, and it is successfully used for the control
of hemorrhage that originates in the bone. However,
animal studies have shown that bone wax can embo-
lize to the lungs
7
and induce a much stronger chronic
inammatory reaction.
8
For this reason, it is rarely
preferred in maxillofacial bone hemorrhages.
In conclusion, the present study shows that bovine
collagen is the most suitable hemostatic agent appli-
cable on peripheral nerves. If the preference for the
hemostasis is oxidized regenerated cellulose, patients
must be informed of the possibility of temporary
sensorial disturbances.
References
1. Loescher AR, Robinson PP: The effect of surgical medicaments
on peripheral nerve function. Br J Oral Maxillofac Surg 36:327,
1998
2. Finn MD, Schow SR, Schneiderman ED: Osseous regeneration in
the presence of 4 common haemostatic agents. J Oral Maxillofac
Surg 52:608, 1992
3. Brodbelt AR, Miles JB, Foy PM, et al: Intraspinal oxidised cellu-
lose causing delayed paraplegia after thoracotomyA report of
three cases. Ann R Coll Surg Engl 84:97, 2002
4. Iwabuchi S, Koike K, Okabe T, et al: Iatrogenic paraplegia
caused by surgicel used for hemostasis during a thoracotomy
Report of a case. Jpn J Surg 27:969, 1997
5. Nagamatsu M, Podratz J, Windebank AJ, et al: Acidity is involved
in the development of neuropathy caused by oxidised cellulose.
J Neurol Sci 146:97, 1997
6. Dutton JJ, Tse DT, Anderson RL: Compressive optic neuropathy
following use of intracranial oxidised cellulose hemostat. Oph-
thalmic Surg 14:487, 1983
7. Robicsek F, Masters TN, Litman L: The embolization of bone
wax from sternotomy incisions. Ann Thorac Surg 31:357, 1981
8. Solheim E, Pinholt EM, Bang G: Effect of local hemostatics on
bone induction in rats: A comparative study of bone wax, brin-
collagen paste, and bioerodible polyorthoester with and without
gentamicin. J Biomed Mater Res 26:791, 1992
634 THE EFFECTS OF HEMOSTATIC AGENTS ON PERIPHERAL NERVE FUNCTION