Effects of cerebrospinal fluid drainage in the prevention of paraplegia after thoracic aorta cross-clamping in a canine model**
(Portuguese
PDF version)
Célio Teixeira Mendonça*
*
PhD, Universidade Federal do Paraná. Post-doctoral fellowship at the
Medical University of South Carolina, USA. Vascular surgeon,
Hospital Nossa Senhora das Graças, Hospital VITA, and Hospital Universitário
Cajuru - Pontifícia Universidade Católica do Paraná (PUC-PR), Curitiba,
PR, Brazil.
** This
work was performed at Instituto de Pesquisa e Cirurgia Experimental
Dr. Egas P. Izique, Universidade Federal do Paraná and it was
awarded with Eduardo C. Palma Prize during the XXIV Latin American
Congress of the International Society for Cardiovascular Surgery and
XVI Encontro Paulista de Cirurgia Vascular.
Correspondence:
Célio Teixeira Mendonça
Rua Visconde do Rio Branco, 1717, 3 andar
CEP 80420-210 - Curitiba - PR, Brazil
Tel.: + 55 (41) 322.5422
Fax: + 55 (41) 3026.3399
E-mail: celiotm@uol.com.br
ABSTRACT
Objective:
To determine if the cerebrospinal fluid drainage could increase
the spinal cord perfusion pressure, and decrease the incidence of
paraplegia after thoracic aorta cross-clamping in dogs, as well
as to study the correlation between the spinal cord perfusion pressure,
the neurologic status of the animals and the degree of histologic
injury to their spinal cords.
Method: Group I animals (n = 6) had a left thoracotomy without
thoracic aorta cross-clamping; Group II animals (n = 6) had a left
thoracotomy with thoracic aorta cross-clamping, and Group III animals
(n = 6) had a left thoracotomy with cerebrospinal fluid drainage
followed by thoracic aorta cross-clamping.
Results: All Group II animals showed evidence of spinal cord
injury; groups I and III animals were neurologically normal (P
= 0.00108). The spinal cord perfusion pressure in Groups I and III
animals was higher than the spinal cord perfusion pressure in Group
II (P = 0.000). The histology of the spinal cords in Groups
I and III animals was normal; in Group II animals, there was infarct
of the motor neurons.
Conclusions:
Cerebrospinal fluid drainage significantly decreased the incidence
of paraplegia after thoracic aorta cross-clamping in this model.
This protective effect was due to the reduction in the cerebrospinal
fluid pressure that caused an increase in the spinal cord perfusion
pressure.
Key
words: paraplegia, drainage, cerebrospinal fluid, thoracic aorta.
Palavras-chave: paraplegia, drenagem, líquido cérebro-espinhal,
aorta torácica.
J
Vasc Br 2004;3(3):181-90
Paraplegia
continues to be a devastating complication after surgical treatment
of thoracic aortic aneurysms and thoracoabdominal aneurysms. The incidence
of this complication may range from 6.5 to 40%, depending on the length
of the aortic segment involved, the presence of dissection or rupture,
the occurrence of peroperative hypotension, and the duration of aortic
occlusion.1-8 Due to all
these factors, there is an increasing interest in developing clinically
relevant experimental methods in order to protect the spinal cord under
these circumstances. Despite the advent of endoluminal repair of thoracic
aortic aneurysms, paraplegia still occurs.9,10
In the majority of cases, paraplegia seems to be directly related to
the decrease of blood flow to the spinal cord during aortic cross-clamping.
Various methods have been employed to prevent spinal cord ischemia after
aortic occlusion, including shunts and aortofemoral bypasses,11
reimplantation of intercostals arteries,4,5
improvement of surgical techniques to reduce aortic occlusion time and
hypothermia.12 Although most of these
techniques may seem beneficial, none of them, so far, consistently prevented
the occurrence of paraplegia in patients submitted to thoracoabdominal
aortic reconstruction.
Also, paraplegia
may be caused or aggravated by reperfusion, which is a complex mechanism
of tissue injury. During ischemia, membrane injury and cell enzyme dysfunction
(xanthine-oxidase system) begin and cell swelling occurs. During reperfusion,
the cell membrane and enzymes suffer abrupt injuries, caused by the
appearance of superoxide and hydroxyl free radicals. Calcium ion rapidly
penetrates the cells (being exchanged for a sodium ion), while acidosis
is promptly corrected. The enzymes that are responsible for calcium
extraction from the cells work inadequately and sodium keeps on entering
at every depolarization, probably contributing to additional calcium
entrance. White cells are then retained by edematous endothelial cells,
causing focal ischemia and additional injury by forming new free radicals.13
Although
dysfunction in the spinal cord occurs in the majority of patients, during
or immediately after aortic cross-clamping, some patients develop delayed
paraplegia, which manifests itself between the first and the third postoperative
day. The cause of this phenomenon is still unclear and its occurrence
has been attributed to postoperative hypotension, embolization, anterior
spinal artery thrombosis or occlusion of the intercostals arteries reimplanted
into the aortic graft. Ackerman & Traynelis14
showed that cerebrospinal fluid (CSF) drainage might, in some cases,
revert delayed paraplegia.
Previous
studies have suggested a relationship between the cerebrospinal fluid
pressure (CSFP) and spinal cord ischemia during the aortic cross-clamping
time.15 Conceptually, the spinal cord
perfusion pressure (SCPP) during aortic occlusion is equal to the arterial
pressure distal to the site of the aortic cross-clamping (or femoral
artery pressure = FAP) subtracted from the CSFP (SCPP = FAP - CSFP).2,16
Maneuvers
that increase the SCPP during aortic occlusion, either increasing distal
arterial pressure at the aortic cross-clamping site or decreasing the
CSFP, may, theoretically, protect the spinal cord from ischemic injury
that occur during the time period when the thoracic artery remains occluded.
This study,
therefore, aims at: a) determining whether CSF drainage may increase
the SCPP and prevent the occurrence of paraplegia after thoracic aorta
cross-clamping in a canine model, and b) correlating the SCPP to the
neurological state of the animals and the degree of histological injury
of their spinal cords.
MATERIAL
AND METHODS
Experimental
design
This study
has received the approval from the Comitê de Bioética
de Pesquisa em Animais (Bioethics Committee for Animal Research)
of Universidade Federal do Paraná (UFPR). Eighteen male mongrel
dogs with an average weight between 7,5 and 15 kg were used. After the
end of the preoperative observation period (7 days), the animals were
randomly distributed into three groups:
-
Group I (n = 6): animals that underwent left thoracotomy without thoracic
aorta cross-clamping.
- Group II (n = 6): animals that underwent left thoracotomy with
thoracic aorta cross-clamping 1 cm distal to the origin of the left
subclavian artery, during 60 minutes.
- Group III (n = 6): animals that underwent left thoracotomy with
CSF drainage before aortic cross-clamping, and thoracic aorta cross-clamping
1 cm distal to the origin of the left subclavian artery, during 60 minutes.
The animals were kept in an adequate kennel during an observation period
of 7 days. A 12-hour fasting period before the surgery was established
for all animals.
Anesthesia
Thirty
minutes prior to the induction of general anesthesia, subcutaneous pre-anesthetic
drugs, such as chlorpromazine (5 mg/ml, Amplictil®, Rhodia), 0.5
mg/kg, and atropine sulphate (0.25 mg/ml, Atropine Sulphate, Apsen),
0.1 mg/kg were administered.
Anesthesia
was induced with intravenous thiobarbiturate (1 methyl-butyl) ethyl
sodium (Thionembutal®, Abbott) 15 to 30 mg/kg of body weight. Afterwards,
the animals were placed on the surgical table in right lateral decubitus
position. Orotracheal intubation was performed and the dog was ventilated
with pressure ventilator under room air.
During
the rest of the procedure, with the use of a universal vaporizer (Takaoka®),
the anesthesia was maintained with endotracheal halothane (Halothane®,
Hoechst) concentration of 1 to 2%. The animals' skin was disinfected
with a solution of Povidine® (Darrow).
Throughout
the procedure, 5% glucose solution in 0.9% sodium chloride solution
(Darrow®) was administered every hour, in a volume of 20 ml/kg of
body weight.
Soon after
anesthesia induction was performed, each animal received cephalothin
sodium (1 g), administered by intravenous injection as single-dose antibiotic
prophylaxis.
Operation
The sterile
surgical procedure was performed as follows:
- Through an incision on the posterior region of the neck, an 18G
Teflon catheter (A-Cath Tecnobio®) was inserted into the subarachnoid
space. This procedure aimed at monitoring the CSFP obtained by cisterna
magna punction under direct observation, avoiding, therefore, CSF
leakage.
- Through incisions on the left lateral region of the neck and right
groin, arterial wires 18G Teflon catheters (A-Cath Tecnobio®)
were inserted into the left carotid artery and right femoral artery,
in order to monitor the pressure over the carotid artery (CAP or arterial
pressure proximal to the level of the aortic cross-clamping) and the
FAP (or arterial pressure distal to the level of the aortic cross-clamping).
- A thermometer was introduced into the anal opening to control rectal
temperature (RT).
The wires for pressure monitoring were all connected to extension tubes
(extension tube Tecnobio® with Luerlock® rotational connector,
measuring 3.3 mm diameter and 120 cm length) and connected to transducers
in a multichannel cardiac monitor (BESE® - Bio-Engineering of Systems
and Equipments Biomonitor 7) with three invasive pressure channels to
monitor CAP, FAP and CSFP, and one electrocardiogram channel.
A left
thoracotomy was performed at the level of the fifth intercostal space.
The descending thoracic aorta was dissected approximately 1 cm distal
to the origin of the left subclavian artery.
Immediately
before the aortic cross-clamping, the animals from Group III had their
CSF drained by a tube connected to the needle that had been introduced
into the magna cistern to monitor the CSFP. Animals from Groups II and
III received intravenous injection of heparin (sodium heparin, Organon
Teknika®), 100 U/kg) and, five minutes later, aortic clamping 1
cm distal to the origin of the left subclavian artery was performed
for 60 minutes. Sodium bicarbonate was administered at a rate of 20
a 25 mEq for five minutes prior to clamp release to minimize the effects
of metabolic acidosis.
Animals
from Groups II and III had their CAP, PAF, and CSFP measured at intervals
of 20, 10, and 5 minutes prior to the aortic cross-clamping; 5, 10,
20, 30, 40, 50 and 60 minutes during the aortic cross-clamping; and
5, 10, and 20 minutes after clamp release (a total number of 13 measurements).
An equal number of measurements (13) was performed in the animals from
Group I at corresponding intervals (this group did not undergo thoracic
aorta cross-clamping).
At the
end of the procedure, the thoracotomy was closed with polyglactin threads
(Vicryl®, Ethicon). All air was aspirated from the pleural space.
The animals were monitored for 24 hours (paraplegic) or 72 hours (normal
or paretic) for evaluation of their neurological state. After this evaluation
the animals were sacrificed with an intravenous injection (20 ml) of
potassium chloride 19.1%, which caused cardiac arrest followed by respiratory
arrest.
Evaluation
of the neurological state
The neurological
state of the animals was evaluated immediately after they recovered
from anesthesia and 24 or 72 hours postoperative, according to the Tarlov
scale.17
Tarlov
scale:
0 = absence
of lower limb movements
1 = perceptible
lower limb movements
2 = good
capacity of movement in the lower limbs but inability to maintain standing
position
3 = capacity
to stand up and walk with some difficulty
4 = total
recovery
Animals
with score 0 were considered paraplegic and were sacrificed after a
24-hour period of observation, in order to avoid any kind of unnecessary
suffering. Animals with scores 1 to 3 were considered paretic, and animals
with score 4 were considered normal: they were observed for a period
of 72 hours to see if their neurological state would deteriorated (a
phenomena known as delayed paraplegia).14,18
After this period of observation, these animals were also sacrificed.
Histological
analysis
Immediately
after the animals were sacrificed, their spinal cords were removed and
placed in 10% buffered formalin solution for later histological analysis.
Cuts of the lower thoracic and lumbar/sacral spinal cord, from all animals,
were stained with hematoxylin--eosin so that a pathologist could evaluate
the extent of the spinal cord injury and, therefore correlate the degree
of the histological injury of the spinal cord with the neurological
state of the animals and their SCPP. The pathologist had no previous
knowledge of the groups to which the animals belonged.
Statistical
methodology
The response
variables evaluated were FAP, CAP, CSFP, RT, and SCPP. All pressure
values were reported in mmHg ± standard error and the temperatures
were reported in Celsius degrees ± standard error. The design
of this experiment was completely randomized and the factors studied
were the clamping condition of the thoracic aorta and CSF drainage.
These factors presented themselves in three levels: non cross-clamping
condition (Group I), cross-clamping condition (Group II), and CSF drainage
condition followed by cross-clamping (Group III).
In order
to compare the three levels of each factor, a classical Analysis of
Variance (ANOVA) was applied, after observing the following three premises:
independence, homocedasticity, and gaussianity. The last two conditions
were tested by classical methodology (Filiben test for gaussianity of
the residuals of the model and Cochran test for homocedasticity). When
the premises did not occur, the Krushall-Willis procedure was used to
compare the groups.
The statistical
analysis of the neurological state was performed by comparing normal
neurological state versus abnormal neurological state of the animals
from the three groups. Specifically, it was investigated whether the
distribution of the animals into the categories paraplegic, paretic,
and normal was the same in Groups I, II, and III. For this investigation,
2x2 contingency tables were built, associating the categories and the
groups. Chi-square test and Fisher's exact test were employed. The P
unilateral value refers to the Fisher's test.
The data
was computed using the Minitab statistical software.
RESULTS
Temperature
There was
no statistically significant difference between the RT of the animals
belonging to the three groups in the time intervals analyzed in this
study.
Hemodynamic
measurements
Group
I
Animals
from Group I presented minimal variations of CAP, FAP, and CSFP throughout
the experiment. The mean SCPP during the interval correspondent to the
60-minute period of aortic cross-clamping was 95.07 ± 1.62 mmHg
(Figures 1, 2, 3, and 4 and Tables 1, 2, 3, and 4).
 Figure
1 - Variation of CAP in Groups I (dotted line), II (dashed line), and
III (continuous line) during the experiment.
Figure
2 - Variation of FAP in Groups I (dotted line), II (dashed line), and
III (continuous line) during the experiment.
Figure
3 - Variation of CSFP in Groups I (dotted line), II (dashed line), and
III (continuous line) during the experiment.
Figure
4 - Variation of SCPP in Groups I (dotted line), II (dashed line), and
III (continuous line) during the experiment.
Table
1 - Carotid artery pressure values in mmHg (mean ± standard
error) for the three groups during the experiment (from 20 minutes of
pre-clamping to 20 minutes after aorta unclamping)
 |
| Group
|
Time
period |
|
Pre-clamping
|
Post-clamping |
|
|
20
min |
10
min |
5
min |
5
min |
10
min |
20
min |
| I
|
97.3±6.1
|
100.1±5
|
95.6±4.2
|
98.3±1.5
|
97.8±3.9
|
99±5.3 |
| II
|
108±2
|
109.3±1.5
|
107.3±4.6
|
134.8±5.7 |
138.3±5.6
|
145±4.7 |
| III
|
110±4.7
|
108.6±4.5
|
109±1.6
|
148.3±4
|
154±5 |
135.3±6.2 |
|
|
Post-clamping
|
Post-unclamping |
|
|
30
min |
40 min |
50
min |
60
min |
5
min |
10
min |
20
min |
| I
|
99.3±2.2
|
99.3±2.2
|
96.3±1.9
|
99.6±2.6
|
101.1±2.6
|
102.1±4
|
98.1±3.6 |
| II
|
145.3±5.3
|
145.1±6.5
|
146.1±5.7
|
145.5±6.5
|
98±11
|
96.1±11
|
102.5±8.9 |
| III
|
145.3±3.9
|
150.8±4.8 |
146.3±4.4
|
144.1±4.2
|
112±5.8
|
104.1±5.6
|
104.1±6.3 |
|
Table
2 - Anatomical distribution of aneurysms with different diameters
|
|
Group
|
Time
period |
|
|
Pre-clamping |
Post-clamping |
|
|
|
20
min |
10
min |
5 min |
5
min |
10
min |
20
min |
|
I
|
98±5.2 |
99.6±5 |
96.6±4.4 |
99±1.5 |
99.6±4 |
99.8±5.5 |
|
II
|
106.5±2 |
107.3±1.9 |
106.1±4.6 |
23±2.5 |
23.8±2.4 |
26±2.4 |
|
III
|
107.8±4.7 |
108.6±4.1 |
106±2.3 |
22.3±1.1 |
24.6±0.9
|
25±1 |
|
|
|
Post-clamping
|
Post-unclamping |
|
|
|
30
min |
40 min |
50
min |
60
min |
5
min |
10
min |
20
min |
|
I
|
98.8±2.2 |
99.6±2.4 |
97.5±2.3 |
98.6±2.7 |
100.8±2.1 |
102.3±3.9 |
97.3±3.2 |
|
II
|
27.3±2.4 |
27.8±2 |
28.6±1.9 |
29.1±2.1 |
96.3±10 |
96.1±11 |
101.3±8.2 |
|
III
|
26.1±0.9
|
28.6±1.1
|
29.8±1.4 |
29.6±1.4
|
110.1±4.3
|
101.3±4.7
|
102.1±7 |
|
Table
3 - Cerebral spinal fluid pressure values in mmHg (mean ± standard
error) for the three groups during the experiment (from 20 minutes of pre-clamping
to 20 minutes after aorta unclamping
|
| Group
|
Time
period |
|
Pre-clamping
|
Post-clamping |
|
|
20
min |
10
min |
5
min |
5
min |
10
min |
20
min |
| I |
4±0.3 |
4±0.3 |
4±0.3 |
4.1±0.4 |
4±0.5 |
3.6±0.3 |
| II |
5.5±0.5 |
5.6±0.4 |
5.3±0.5 |
9.8±0.7 |
9.8±0.6 |
10.1±0.7 |
| III
|
4.5±0.4
|
4.5±0.4
|
-8.1±0.3
|
-8.3±0.6
|
-8.1±0.7 |
-8.1±0.5 |
|
|
Post-clamping
|
Post-unclamping |
|
|
30
min |
40 min |
50
min |
60
min |
5
min |
10
min |
20
min |
| I |
4.3±0.3 |
4±0.4 |
3.6±0.3 |
3.8±0.4
|
4±0.4 |
3.5±0.3 |
3.5±0.3 |
| II |
10.3±0.7 |
10.3±0.8 |
10.5±0.9 |
10.5±0.8 |
7.8±0.7 |
7.5±0.6 |
7.1±0.6 |
| III
|
-7.6±0.6
|
-7.8±0.7
|
-7.8±0.7
|
-7.3±0.6
|
-7.3±0.8
|
-6.6±0.9
|
-7.1±0.8 |
|
Table
4 - Spinal cord perfusion pressure values in mmHg (mean ± standard
error) for the three groups during the experiment (from 5 to 60 minutes after
aorta cross-clamping)
|
| Group
|
Time
period |
|
Pre-clamping
|
Post-clamping |
|
|
20
min |
10
min |
5
min |
5
min |
10
min |
20
min |
| I |
-------------- |
------------ |
----------- |
94.8±1.4 |
95.6±4.3 |
96.1±5.8 |
| II |
-------------- |
------------ |
----------- |
13.1±2.5 |
14±2.2 |
15.8±2.3 |
| III
|
--------------
|
------------ |
----------- |
30.6±1.5
|
32.8±1.4
|
33.1±1.3 |
|
|
Post-clamping
|
Post-unclamping |
|
|
30
min |
40
min |
50
min |
60
min |
5
min |
10
min |
20
min |
| I |
94.5±2.2 |
95.6±2.1 |
93.8±2.3 |
94.8±2.5 |
------------ |
------------ |
------------ |
| II |
17±2.2 |
17.5±1.9 |
18.1±1.7 |
18.6±1.9 |
------------ |
------------ |
------------ |
| III
|
33.8±1.4 |
36.5±1.6 |
37.6±1.8
|
37±1.8 |
------------ |
------------ |
------------ |
|
Group
II
Figures
1, 2, 3, and 4 and Tables 1, 2, 3, and 4 shows the variations of CAP,
FAP, CSFP, and SCPP. The mean SCPP during the interval correspondent
to the 60-minute period of cross-clamping was 16.33 ± 2.1 mmHg.
Only one animal presented paresis, and its SCPP was 25.42 mmHg.
Group
III
Figures
1, 2, 3, and 4 and Tables 1, 2, 3, and 4 shows the variations of CAP,
FAP, CSFP, and SCPP. The CSFP decreased from the baseline value of 4.5
± 0.4 mmHg to -8,1 ± 0.3 mmHg, immediately after CSF drainage.
An average amount of 11.33 ± 0.71 ml of CSF was taken from the
subarachnoid space of the dogs from Group III, immediately before the
aortic cross-clamping. The mean SCPP during the interval correspondent
to the 60-minute period of the aortic cross-clamping was 34.52 ±
1.52 mmHg.
- "
The CAP of the animals from Group II and III was significantly higher
than the CAP of the animals from Group I, for the time interval that
corresponded to the aortic cross-clamping (from 5 to 60 minutes after
the cross-clamping) (Figure 1).
- The FAP of the animals from Groups II e III was significantly
lower that the FAP of the animals from Group I, for the time interval
that corresponded to the aorta cross-clamping (from 5 to 60 minutes
after cross-clamping) (Figure 2).
- The CSFP of the animals from Group II was significantly higher
than the CSFP of the animals from Group I, which was significantly higher
than the CSFP of the animals from Group III, for the time interval that
corresponded to 5-minute pre-clamping period to 20 minutes after aortic
unclamping (Figure 3).
- The SCPP of the animals from Group I was significantly higher
than the SCPP of the animals from Group III, which was significantly
higher than the SCPP of the animals from Group II, for the time interval
that corresponded to the aortic cross-clamping (from 5 to 60 minutes
after the cross-clamping) (Figure 4).
Evaluation
of the neurological state of the animals
Group I:
all dogs walked normally without any evidence of spinal cord injury,
during the 72-hour period of observation (Tarlov 4).
Group II:
all dogs presented evidences of spinal cord injury: five (83.3%) presented
spastic paraplegia and absence of lower limb movements (Tarlov 0), and
one (16.7%) presented paresia (Tarlov 2).
Group III:
all dogs walked normally without any evidence of spinal cord injury,
during the 72-hour period of observation (Tarlov 4).
Animals from Groups I e III presented better postoperative neurological
evolution when compared to the animals from Group II (P = 0.00108).
Histology of the spinal cord
All animals
of this study (18) had their lower thoracic and lumbar-sacral spinal
cord removed.
From a
neurological point of view, the six animals belonging to Group I were
normal. Optical microscopy showed that the motor neurons located in
the anterior horn on their spinal cords were normal, without any evidence
of ischemic injury to the spinal cord.
Five out
of the six animals from Group II presented with spastic paraplegia and
absence of lower limb movements (Tarlov = 0). Optical microscopy of
their spinal cords showed infarct characterized by gray matter degeneration,
hemorrhage, and death of the motor neurons located in the anterior horn
of the spinal cord (Figure 5). Optical microscopy of the spinal cord
also showed neuronal injury in one animal from Group II presented with
paresia. However, this injury was minor when compared to the injuries
presented by the paraplegic animals.
Figure
5 - Optical microscopy of spinal cord gray matter of one animal (Group
II) presented with paraplegia: note neuronal degeneration in the anterior
horn with ischemia of the surrounded neural tissue (Hematoxylin-eosin, magnification
of 400x).
From a
neurological point of view, all animals from Group III (six animals)
were normal. Optical microscopy showed that the motor neurons located
in the anterior horn of their spinal cords were also normal, showing
no evidence of ischemic injury to the spinal cord (Figure 6).
Figure
6 - Optical microscopy of spinal cord gray matter of one animal (Group
III) showing normal histological aspect (Hematoxylin-eosin, magnification
of 400x).
DISCUSSION
Spinal
cord injury, following thoracoabdominal aortic aneurysm repair, is caused
by many factors including thrombosis or embolization of the critical
intercostal arteries, and permanent disruption of important vessels
in the spinal cord. However, the most important reason for spinal neurological
deficits is prolonged ischemia during aortic cross-clamping, since the
Adamkiewicz artery (or arteria radicularis magna, the principal arterial
blood supply of the spinal cord) is located distally to the clamp that
is used to perform the aortic cross-clamping.
Unfortunately,
there is not, so far, any clinical or experimental method that could
consistently decrease the incidence of this complication. This fact
was one the major reasons that led us to develop an experimental model
to cause neurological injury in a significant number of animals submitted
to aortic cross-clamping. Once having such model, it would be possible
to investigate the efficacy of experimental methods to protect the spinal
cord during aortic cross-clamping (such as CSF drainage) by simply comparing
the incidence of paraplegia in clamped and unprotected dogs (Group II)
to the incidence of paraplegia in clamped and protected dogs with the
use of the method mentioned above (CSF drainage - Group III).
This study
showed that the canine model presented herein was efficient in producing
ischemic injury to the spinal cord: all dogs (100%) from Group II presented
neurological injury, whereas the incidence of neurological injury in
the dogs from Groups I and III was 0%.
The mean
SCPP of Group II dogs was significantly lower than that of Group I dogs
(figure 4) for two basic reasons: FAP decrease (Figure 2) and CSFP increase
(Figure 3) due to the thoracic aortic cross-clamping. This decrease
of the mean SCPP during the aortic cross-clamping period led to death
of the motor neurons in the anterior horn of the spinal cord and neurological
deficit of the animals from Group II.
A major
factor that may be related to this decrease in SCPP is the increase
in CSFP that occurs during the aortic cross-clamping. Experimental studies
show an increase in CSFP during the aortic cross-clamping, ranging from
30 to 100% of the baseline values.19,20
This study
showed that, in this canine model, CSF drainage was efficient, significantly
decreasing the incidence of paraplegia after thoracic aorta occlusion.
The protective effect was due to a decrease in CSFP of Group III animals
(Figure 3), with a consequent increase of SCPP (Figure 4). The CSF drainage,
performed prior to the aortic cross-clamping, increased the mean SCPP
from 16.33 ± 2.1 mmHg for Group II dogs to 34.52 ± 1,52
mmHg for Group III dogs (Figure 4); this increase in the mean SCPP during
the aortic cross-clamping maintained the motor neurons viable and avoided
the occurrence of paraplegia in the dogs from Group III. After the animals
from Group III had their CSF drained, the liquor pressure turned negative
(Figure 3), creating a vacuum effect within the subarachnoid space.
This fact may have increased the capillary blood flow in the spinal
cord, during the aortic cross-clamping, protecting the cord from ischemic
injury, as described by Aadahl.21
Based on
the data provided by our study, CSF monitoring and drainage seems to
be especially attractive for the management of patients presented with
large aneurysms of the thoracoabdominal aorta that requires surgical
treatment. Although CSF drainage do not restore the SCPP to its normal
levels (Figure 4), it causes an increase in the SCPP and, consequently,
an increase of the spinal cord tolerance to the ischemia caused by thoracic
aorta cross-clamping, allowing more time for the surgical procedure
and, perhaps, decreasing the incidence of postoperative paraplegia in
a selected group of patients.
CONCLUSIONS
CSF drainage
showed to be efficient by significantly decreasing the rate of paraplegia
after thoracic aorta occlusion in dogs. The mean SCPP had a good correlation
with the neurological state of the animals observed 24 to 72 after the
experiment and with the level of histological injury to their spinal
cords.
REFERENCES
1.
Stolf NA, Pego Fernandes PM, Bittencourt D, et al. Thoracic aorta aneurysms
and dissections. Surgical treatment. Arq Bras Cardiol 1983;41:221-6.
2. Hollier LH. Protecting the brain and spinal cord.
J Vasc Surg 1987;5:524-8.
3. Cambria RP, Clouse WD, Davison JK, et al. Thoracoabdominal
aneurysm repair: results with 337 operations performed over a 15-year
interval. Ann Surg 2002;236:471-9.
4. Livesay JJ, Cooley DA, Ventemiglia RA, et al. Surgical
experience in descending thoracic aneurysmectomy with and without adjuncts
to avoid ischemia. Ann Thorac Surg 1985;39:37-45.
5. Hollier LH, Money SR, Naslund TC, et al. The risk
of spinal cord dysfunction in patients undergoing thoracoabdominal aortic
replacement. Am J Surg 1992;164:210-14.
6. Crawford ES, Waler HSJ III, Saleh SA, et al. Graft
replacement of aneurysm in descending thoracic aorta: results without
bypass or shunting. Surgery 1981;89:73-85.
7. Svensson LG, Crawford ES, Hess KR, et al. Experience
with 1509 patients undergoing thoracoabdominal aortic operations. J
Vasc Surg 1993;17:357-70.
8. Zvara DA. Thoracoabdominal aneurysm surgery and the
risk of paraplegia: contemporary practice and future directions. J Extra
Corpor Technol 2002;34:11-17.
9. Dake MD, Miller DC, Semba CP, et al. Transluminal
placement of endovascular stent-grafts for the treatment of descending
thoracic aortic aneurysms. N Eng J Med 1994;331:1729-34.
10. Scharrer-Pamler R, Kotsis T, Kapfer X, et al. Complications
after endovascular treatment of thoracic aortic aneurysms. J Endovasc
Ther 2003;10:711-18.
11. Estrera AL, Rubenstein FS, Miller CC, et al. Descending
thoracic aortic aneurysm: surgical approach and treatment using the
adjuncts cerebrospinal fluid drainage and distal aortic perfusion. Ann
Thorac Surg 2001;72:481-6.
12. Cambria RP, Davison JK, Carter C, et al. Epidural
cooling for spinal cord protection during thoracoabdominal aneurysm
repair: A five-year experience. J Vasc Surg 2000;31:1093-102.
13. McCord JM. Oxygen-derived free radicals in postischemic
tissue injury. N Engl J Med 1985;312:159-63.
14. Ackerman LL, Traynelis VC. Treatment of delayed-onset
neurological deficit after aortic surgery with lumbar cerebrospinal
fluid drainage. Neurosurgery 2002;51:1414-21.
15. Coselli JS, Lemaire SA, Koksoy C, et al. Cerebrospinal
fluid drainage reduces paraplegia after thoracoabdominal aortic aneurysm
repair: results of a randomized clinical trial. J Vasc Surg 2002;35:631-9.
16. McCullough JL, Hollier LH, Nugent M. Paraplegia
after thoracic aortic occlusion: influence of cerebrospinal fluid drainage.
J Vasc Surg 1988;7:153-60.
17. Tarlov IM. Spinal cord compression. Mechanisms
of paralysis and treatment. 1st ed. Springfield, Illinois: Charles C.
Thomas; 1957.
18. Moore WM Jr, Hollier L. The influence of severity
of spinal cord ischemia in the etiology of delayed-onset paraplegia.
Ann Surg 1991;213:427-31.
19.
Blaisdell FW, Cooley DA. The mechanism of paraplegia after temporary
thoracic aortic occlusion in its relationship to spinal fluid pressure.
Surgery 1962;51:351-5.
20.
Oka Y, Miyamoto T. Prevention of spinal cord injury after cross-clamping
of the thoracic aorta. Jpn J Surg 1984;14:159-62.
21.
Aadahl P, Saether OD, Stenseth R, et al. Microcirculation of the spinal
cord during proximal aortic cross-clamping. Eur J Vasc Surg 1990;4:5-10.
|
