J Neurosurg 47:525-531, 1977
Experimental cerebral concussion Part 1: An electron microscopic study L o u i s BAKAY, M.D., JOSEPH C. LEE, M . D . , PH.D., GEORGE C. LEE, PH.D., AND JIUN-RONG PENG, M.D.
Departments of Neurosurgery, Anatomical Sciences, and Civil Engineering, State University of New York at Buffalo, Buffalo, New York v' Cerebral concussion was produced in rats by an iron pendulum hitting the external occipital protuberance. This resulted in loss of consciousness lasting from 3 to 10 minutes with prompt recovery and no focal neurological signs. The energy absorbed by the head at the impact was calculated to be about 1450 gm/cm. Light microscopic survey showed only minor pathological changes. However, electron microscopic observation revealed considerable alteration which began at 30 minutes, reached a peak at 1 hour, and disappeared at 24 hours after concussion. The salient changes included severe swelling of the neuronal mitochondria at the point of impact (occipital cortex), and extracellular edema at the site of contre coup (frontal lobe). Topographically, the most severe alteration was seen in structures at the craniospinal junction (medulla oblongata and upper cervical cord), consisting of both mitochondrial and edematous changes. Although there was no visible opening of the capillary interendothelial junctions, extravasated ferritin particles were accumulated in the edematous regions, indicating a transient increase in the permeability of the blood-brain barrier. KEY WORDS
mitochondria
cerebral concussion 9 electron microscope
9
EREBRAL concussion has been defined as a clinical entity characterized by a reversible loss of consciousness of short duration without localizing neurological features. There has not been any anatomical substrate associated with concussion in the f o r m of detectable pathology, 5,6 nor has there been any evidence of physical d a m a g e to the brain, xe H o w e v e r , the permeability of the blood capillaries in the concussed brain is increased for a short duration. The lack of a n a t o m i c a l changes in the brain became a conditio sine qua non, a matter of great definitional importance. Upon this condition rests the differentiation between concussion and contusion. Yet all the
c
J. Neurosurg. / Volume 4 7 / October, 1977
9
cerebral edema
9
cardinal statements of the past referring to this point were based on examination by light microscopy, a method that reveals little a b o u t finer changes of the nervous system. W e decided to study the m a t t e r by electron m i c r o s c o p y . A n i m a l s were used as experimental subjects because h u m a n brains from autopsies with known and graded times from the onset of cerebral concussion are virtually nonexistent. Material and Methods
The experiments were carried out in 12 young adult rats of Sprague-Dawley strain, weighing 250 to 350 gm. Litter m a t e s were 525
L. Bakay, J. C. Lee, G. C. Lee and J. R. Peng
FIG. 1. Pendulum delivering impact to a rat lying on the horizontal wooden board with its head resting on the end piece at 45 ~ to the horizontal board. used as controls. The experimental animals were subjected to cerebral concussion by a specially designed iron pendulum (Fig. 1). The device provided a known angle for the pendulum to hit the head and a constant position for the rat to rest its head without being held in position by any means. The head moved under the impact and consequently was a true replica of the human head receiving a blow. Under light ether anesthesia, it was placed on the horizontal wooden board with its head resting on an adjustable end piece at 45 ~ to the horizontal board (Fig. 1). The head and nape were shaved to minimize the loss of concussion impact. As soon as the rat showed a slight sign of waking up, the iron pendulum was released, and hit the head from a position 90 ~ to the vertical line of gravity. At the time of impact, the head moved slightly forward, and the pendulum swung further for a dis526
tance of 35 ~ from the vertical line. The lower end of the pendulum was marked with colored chalk and its point of impact could be identified in relation to the external occipital p r o t u b e r a n c e . R a t s not hit at the protuberance were not included in this report, and those that showed signs of hemorrhagic contusion were discarded. The control animals were left on the wooden board to wake up without being struck by the pendulum. The dynamic force of the impact was determined by trial and error in our pilot experiments. There was no loss of consciousness or any other evidence of concussion in the rats when hit by the pendulum dropped from a position 45 ~ to the vertical plane. When the pendulum was dropped from a horizontal position of 90 ~ to the line of gravity, the rats showed transitory unconsciousness for 3 to 10 minutes. Thus, the pendulum swung from 90 ~ to the vertical plane was selected for the experiments on 12 rats, and the energy absorbed by the head was calculated. The energy absorbed by the rat's head was determined from the difference in potential energy corresponding to the initial and final positions of the pendulum. If the initial angle and the angle after impact are Ao and At, respectively, the difference in potential energy ( E o - E~) can be calculated as: Eo - Er = W
R9 (cos At - cos Ao),
where W is the weight of the pendulum and R is the distance from the center of gravity to the support along the pendulum (Fig. 2). Friction energy loss o f the testing system was determined by measuring the difference between the initial and final angles in a "freeswing" situation. This angle due to friction loss was subtracted from the initial angle in computing the energy absorption of the animal. In the experiments conducted, the initial angle was always set at 90 ~. This had an angle of friction loss of 5 ~ in the system. Therefore an initial angle of 85 ~ was used. Other values used in the computation were: W = 1505 gm and R = 13.1 cm. The energy absorption curve is given in Fig. 2. The dotted line shows that for a 35 ~ angle after impact, the energy absorbed is approximately 1450 g m / c m . The 12 rats were killed at the following times after concussion: two at 30 minutes, four at 1 hour, two at 2 hours, two at 6 hours, J. Neurosurg. / Volume 47 / October, 1977
Experimental cerebral concussion and two at 24 hours. The rats were anesthetized by an intraperitoneal injection of a 3.5% chloral hydrate solution (1 ml/100 gm body weight). Cadmium-free ferritin in the amount of 0.3 to 0.5 ml (4.5 mg/ml) was injected into the tail vein, 5 to 15 minutes before death. The animals were killed by perfusion of 20 to 30 ml of warm phosphate buffer (0.1 M at 37 ~ C) through the ascending aorta, followed by 100-to 120 ml of cold 4% glutaraldehyde (in 0. I M phosphate buffer at 4 ~ C). Immediately after the perfusionfixation the cranium was opened, and the brain and upper cervical spinal cord removed. The following specimens were collected: the right and left frontal cortex and subcortical white matter, the right and left occipital cortex and subcortical white matter, the cerebellar cortex and white matter, both superior colliculi, the tegmentum of the medulla oblongata, and the gray and white matter of the first two segments of the cer,r cord. All specimens were postfixed in cold osmium tetroxide for 90 minutes, dehydrated in graded ethanols, and embedded in Maraglas mixture. Thick (1 um) and thin (60 nm) sections were cut from each specimen on a PorterBlum MT-2 ultramicrotome.* The thick sections were stained with 0.1% toluidine blue for light microscopy. The thin sections were stained with uranyl acetate and lead citrate for examination by a JEOLCO 100B electron microscope.t Results
All the rats were unconscious for 3 to 10 minutes after the concussion. During this time the corneal and light reflexes were absent and the animals did not respond to painful stimulation. After they awakened, they remained sluggish for 15 to 20 minutes. Thereafter they behaved and fed normally, and showed no evidence of focal or generalized neurological abnormality. On gross examination the brains of the concussed rats were normal. In particular there was no evidence of swelling or hemorrhages. *Porter-Blum MT-2 ultramicrotome manufactured by Ivan Sorvall, Inc., Norwalk, Connecticut. tElectron microscope supplied by Jeol, Inc., 477 Riverside Avenue, Medford, Massachusetts. J. Neurosurg. / Volume 4 7 / October, 1977
t
!o o
90*~
90*
,;,' -%.+
k':'G:iV'-.--__i_~"~"W*~
//
_ t-inal ~//
i
~t
DATUM
/
~ot.6 ~t.4 21.2 g
0.1 5~
O~
35"
40 ~
45"
50 ~
55 ~
Angle offer impact
FIG. 2. Upper."Diagram showing the initial and final positions of the iron rod with reference to the line of gravity where the rat's head received the impact. Lower."Graph representing the variation in energy absorption as a function of the angle after impact, for an initial angle of 90*. For a postimpact angle of 35 ~ the energy absorbed is 1450 gm/cm. The curve has been adjusted for friction energy loss of the testing system. Light microscopy revealed slight edema in the frontal white matter, the medulla, and upper cervical cord. A few neurons in the occipital cortex and a considerable number of neurons in the cervical cord appeared slightly swollen but, in contradiction to previous findings,* there was no evidence of chromatolysis in the nerve cells. In general, the central nervous system specimens showed only minor changes under the light microscope. Electron microscopy demonstrated significant changes in the brain ultrastructure of the concussed rats; the abnormalities showed great regional variation. 527
L. Bakay, J. C. Lee, G. C. Lee and J. R. Peng
FIG. 3. Left: Occipital cortex of control rat revealing intact neuronal mitochondria and narrow extracellular spaces. X 15,000. Right: Occipital cortex in an experimer]tal rat 1 hour after concussion. Many neuronal mitochondria (m) are moderately swollen, and their cristae are pushed to the periphery. The ribosomes (R), the endoplasmic reticulum (Er) (Nissl substance), the Golgi complexes (G), and the nucleus (N) are normal. X 5000.
The brains of control animals revealed a well known ultrastructural picture of intact mitochondria and narrow extracellular spaces (Fig. 3 left). By comparison, in the occipital cortex of the concussed rats, the point of impact, m a n y mitochondria of the neurons were swollen (Fig. 3 right). The mitochondrial cri~tae were pushed to the periphery by excess fluid. T h e m i t o c h o n d r i a of glial cells remained normal. The nucleus, Nissl substance, and Golgi complex of the neurons were not altered, nor was there any evidence of chromatolysis. There was no edema in the occipital cortex or white matter. Changes in the cerebellar cortex were similar to those found in the occipital lobe. The superior colliculi remained entirely intact. The most severe damage occurred in the medulla oblongata and the upper cervical cord. Here numerous neuronal mitochondria were found to be greatly swollen (Fig. 4 left). In some instances the cristae could barely be 528
seen because they were pushed against the outer membrane. There was slight enlargement of the extracellular space in the neuropil. The white matter, on the other hand, showed considerable extracellular edema (Fig. 4 right). The axons revealed slight overhydration. There was some ruffling of the myelin sheaths. A great concentration of gliofilaments was present in the processes of the fibrous astrocytes. The frontal cortex, the site of the contre coup, revealed minor changes in terms of slight swelling of the perivascular astrocytic processes and mild enlargement of the extracellular space (Fig. 5 left). Although the "tight junction" of the capillary endothelium was not widened, numerous ferritin particles were present in the basal lamina of the capillaries and venules. A few ferritin particles were even seen in the extracellular space, indicating increased permeability of the blood-brain barrier. There was occasional
J. Neurosurg. / Volume 47 / October, 1977
Experimental cerebral concussion
FIG. 4. Left: Neuropil of the gray matter of the upper cervical cord 1 hour after concussion. Numerous neuronal mitochondria (m) are swollen, and most of their cristae can be seen only over part of the periphery. Other subcellular organelles and synapses (s) appear intact. The extracellular space (e) is slightly enlarged. X 3000. Right: White matter of the upper cervical cord 1 hour after concussion. The extracellular space (e) is greatly enlarged. The myelin sheaths (Ms) of some nerve fibers show slight rufflings. The axons (Ax) appear somewhat overhydrated. A great concentration of gliofilaments (Gf) is evident in the processes of fibrous astrocytes. • 8500.
FIG. 5. Left: Frontal cortex 1 hour after concussion. The perivascular astrocytic processes (A) are moderately swollen. The extracellular space (e) in the neuropil is slightly enlarged. A few mitochondria (m) appear partially hydrated. The "tight junction" (T) of the endothelial cells is not widened. Numerous ferritin particles can be seen in the basal lamina (B) of a capillary, and some in the swollen astrocytic processes and the enlarged extracellular space (arrows). L = lumen of the capillary. X 4000. Right: Subcortical white matter of the frontal lobe 1 hour after concussion. The extracellular space (e) is enlarged. The axons (Ax) and their mitochondria (m) appear intact. The myelin sheaths of some nerve fibers show rufflings (Ms). • 3250.
J. Neurosurg. / Volume 4 7 / October, 1977
529
L. Bakay, J. C. L e e , G. C. L e e a n d J. R. P e n g mild swelling of the neuronal and glial mitochondria. In the frontal white matter, there was a definite extracellular edema as evidenced by the enlarged extracellular spaces (Fig. 5 right). Some ruffling of the laminae of the myelin sheaths was also observed. These alterations were already observed 30 minutes after the concussion but in a lesser degree of severity. They reached their peak at 1 hour, declined afterward, and virtually disappeared by 24 hours. The severe changes of the mitochondria were completely reversible within 24 hours, while a slight enlargement of the extracellular space of the white matter in the frontal lobes and upper cervical cord remained. The same changes were observed in the brains of rats sacrificed in the same period of time after concussion. Discussion
Previous studies of experimental brain concussion were performed on animals without a known force of the impact on the skull or with a calculated force but after exposure of the brain. 8,9 More accurate data were obtained from the experiments of Ommaya, et al. ~~ We believe that the simple adjustable pendulum used in our experiments, exerting its force on the intact skull, comes closest to the situation prevailing in the average head injury in man that results in cerebral concussion, because the transient unconsciousness is followed by complete recovery without any focal neurological signs. It is still amenable to the calculation of the force of impact in dynamic terms, readily providing us with the actual energy absorbed by the head at impact. Our observations confirm the statement made by Brodersen 3 that the severity of clinical symptoms is proportional to the energy absorbed by the head on impact. Our findings also indicate that the major damage occurs in the frontal region and at the craniospinal junction, while the mesencephalon and the caudal diencephalon are the least vulnerable. This is comparable with the findings by Ommaya and Gennarelli. 1~ From a mechanical point of view, the changes are caused by contre coup as far as the frontal lobes are concerned and stretchshearing effects at the craniospinal junction. In summarizing the effect of concussion at the ultrastructural level, mitochondrial changes occur in the neurons of the occipital 530
cortex, the region of the impact (Fig. 3 right). The changes in the structure of the neuronal mitochondria are even more severe in the medulla and upper cervical cord, areas somewhat remote from the impact but subject to traumatic stretch and shearing (Fig. 4). However, it should be pointed out that swelling of neuronal mitochondria is not specific to brain concussion. It was observed in a variety of neurological diseases including acute and chronic hypoxia and radiation damage, to name a f e w . 2,15,17 Interestingly enough, edema does not develop in the impact region. It occurs in areas farther away, such as the frontal lobes and the craniospinal junction. The fact that the mesencephalon, as represented by the superior colliculi, remains intact can be explained by the basic mechanical principles involved. The skull and its contents can be viewed as an oval globe, round in cross section and oval in horizontal section. Any force hitting the skull at one point can be resolved in two component forces that meet at the farthest opposite point, giving rise to a contre-coup effect and to a shearing action at the craniospinal junction, s Since the superior colliculi are close to the center of globe, they receive zero force from any impact to the periphery of the ovoid mass. The localized cerebral edema caused by concussion is of the vasogenic type and reflects the transient increase in the permeability of the blood-brain barrier as shown by the extravasation of ferritin. Similar findings were reported earlier with the use of methylene blue, TM fluorescein, 1~ Evans blue-albumin, 1~ and W2.4 In most experiments utilizing vital dyes, increased permeability was restricted to the first 24 hours. Previous experiments by Bakay, et al., 1 using nuclear magnetic resonance technique revealed that the edema fluid in trauma consists of two components. One is motionally restricted, " b o u n d " water, the other motionally unrestricted, "bulk" water. The rapid development, shift, and disappearance of excess water in concussion must be attributed to the fast moving water component. The sudden onset of mitochondrial swelling and its prompt disappearance in 24 hours is probably not caused by a complete disappearance and reappearance of mitochondrial enzymes. A more likely explanaJ. Neurosurg. / Volume 47 / October, 1977
Experimental cerebral concussion tion is a sudden increase in the permeability of the mitochondrial outer m e m b r a n e . The inward flow of excess fluid results in a swelling of the organelle, which pushes the cristae to the p e r i p h e r y and dilutes the m i t o chondrial enzymes. H a m b e r g e r and Rinder 9 found a brief increase in the succinoxidase activity and concluded that concussion activated the existent enzyme in the neuronal mitochondria rather than increasing the synthesis of the enzyme. Detailed histochemical studies now in progress in our laboratory might shed light on this aspect of cerebral concussion. These morphological investigations do not by t h e m s e l v e s explain the loss of consciousness in concussion. H o w e v e r , the edematous and mitochondrial changes of the medulla and upper cervical cord m a y explain such clinical findings as respiratory distress, cardiovascular alterations, and slowing of cerebral circulation? ,7,x~ References
1. Bakay L, Kurland R J, Parrish RG, et al: Nuclear magnetic resonance studies in normal and edematous brain tissue. Exp Brain Res 23:241-248, 1975 2. Bakay L, Lee JC: The effect of acute hypoxia and hypercapnia on the ultrastructure of the central nervous system. Brain 91:697-706, 1968 3. Brodersen P: Posthypoxic and traumatic coma: aspects of coma mechanisms, in Ingvar DH, Lassen NA (eds): Brain Work. The Coupling of Function, Metabolism and Blood Flow in the Brain. Copenhagen: Munksgaard, 1975, pp 269-279 4. Cassen B, Neff R: Blood-brain barrier behavior during temporary concussion. Am J Physiol 198:1296-1298, 1960 5. Denny-Brown D: Brain trauma and concussion. Arch Neurnl 5:1-3, 1961 6. Denny-Brown D, Russell WR: Experimental cerebral concussion. Brain 64:93-164, 1941 7. Gronwall D, Wrightson P: Cumulative effect of concussion. Lancet 2:995-997, 1975 8. Gurdjian ES: Recent advances in the study of the mechanism of impact injury of the head - A summary. Clin Neurosurg 19:1-42, 1972
J. Neurosurg. / Volume 47 / October, 1977
9. Hamberger A, Rinder L: Experimental brain concussion. J Neuropathol Exp Neurol 25:68-75, 1966 10. Ommaya AK, Gennarelli TA: Cerebral concussion and traumatic unconsciousness. Correlation of experimental and clinical observations on blunt head injuries. Brain 97:633-654, 1974 11. Ommaya AK, Rockoff SD, Baldwin M, et al" Experimental concussion. A first report. J Neurosurg 21:249-264, 1964 12. Quadbeck G, Randerath K: Enfluss der Gehirnerschfitterung auf den Obertritt yon Methylenblau aus dem Blut in das Gehirn der Katze. Z Naturforsch 10B:168-173, 1955 13. Rinder L, Olsson Y: Studies on vascular permeability changes in experimental brain concussion. I. Distribution of circulating fluorescent indicators in brain and cervical cord after sudden mechanical loading of the brain. Acta Neuropathoi 11:183-200, 1968 14. Rinder L, Olsson Y: Studies on vascular permeability changes in experimental brain concussion. II. Duration of altered permeability. Acta Neuropathol 11:201-209, 1968 15. Roizin L, Orlovskaja D, Liu JC, et al: Acid-/3glycerophosphatase reaction products in the central nervous system mitochondria following x-ray irradiation. J Histocbem Cytocbem 23:402-410, 1975 16. Ward AA Jr: The physiology of concussion. Clin Neurosurg 12:95-111, 1966 17. Yu MC, Bakay L, Lee JC: Ultrastructure of the central nervous system after prolonged hypoxia. I. Neuronal alterations. Acta Neuropathol 22:222-234, 1972
This paper was presented at the Annual Meeting of the American Association of Neurological Surgeons in Toronto, Canada, April 24-28, 1977. This work was supported by the Buswell Foundation and, in part, by the Research Institute of Alcoholism, Buffalo, New York. Address reprint requests to: Louis Bakay, M.D., Department of Neurosurgery, Clinical Center, 462 Grider Street, Buffalo, New York 14215.
531
Experimental cerebral concussion Part 1: An electron microscopic study L o u i s BAKAY, M.D., JOSEPH C. LEE, M . D . , PH.D., GEORGE C. LEE, PH.D., AND JIUN-RONG PENG, M.D.
Departments of Neurosurgery, Anatomical Sciences, and Civil Engineering, State University of New York at Buffalo, Buffalo, New York v' Cerebral concussion was produced in rats by an iron pendulum hitting the external occipital protuberance. This resulted in loss of consciousness lasting from 3 to 10 minutes with prompt recovery and no focal neurological signs. The energy absorbed by the head at the impact was calculated to be about 1450 gm/cm. Light microscopic survey showed only minor pathological changes. However, electron microscopic observation revealed considerable alteration which began at 30 minutes, reached a peak at 1 hour, and disappeared at 24 hours after concussion. The salient changes included severe swelling of the neuronal mitochondria at the point of impact (occipital cortex), and extracellular edema at the site of contre coup (frontal lobe). Topographically, the most severe alteration was seen in structures at the craniospinal junction (medulla oblongata and upper cervical cord), consisting of both mitochondrial and edematous changes. Although there was no visible opening of the capillary interendothelial junctions, extravasated ferritin particles were accumulated in the edematous regions, indicating a transient increase in the permeability of the blood-brain barrier. KEY WORDS
mitochondria
cerebral concussion 9 electron microscope
9
EREBRAL concussion has been defined as a clinical entity characterized by a reversible loss of consciousness of short duration without localizing neurological features. There has not been any anatomical substrate associated with concussion in the f o r m of detectable pathology, 5,6 nor has there been any evidence of physical d a m a g e to the brain, xe H o w e v e r , the permeability of the blood capillaries in the concussed brain is increased for a short duration. The lack of a n a t o m i c a l changes in the brain became a conditio sine qua non, a matter of great definitional importance. Upon this condition rests the differentiation between concussion and contusion. Yet all the
c
J. Neurosurg. / Volume 4 7 / October, 1977
9
cerebral edema
9
cardinal statements of the past referring to this point were based on examination by light microscopy, a method that reveals little a b o u t finer changes of the nervous system. W e decided to study the m a t t e r by electron m i c r o s c o p y . A n i m a l s were used as experimental subjects because h u m a n brains from autopsies with known and graded times from the onset of cerebral concussion are virtually nonexistent. Material and Methods
The experiments were carried out in 12 young adult rats of Sprague-Dawley strain, weighing 250 to 350 gm. Litter m a t e s were 525
L. Bakay, J. C. Lee, G. C. Lee and J. R. Peng
FIG. 1. Pendulum delivering impact to a rat lying on the horizontal wooden board with its head resting on the end piece at 45 ~ to the horizontal board. used as controls. The experimental animals were subjected to cerebral concussion by a specially designed iron pendulum (Fig. 1). The device provided a known angle for the pendulum to hit the head and a constant position for the rat to rest its head without being held in position by any means. The head moved under the impact and consequently was a true replica of the human head receiving a blow. Under light ether anesthesia, it was placed on the horizontal wooden board with its head resting on an adjustable end piece at 45 ~ to the horizontal board (Fig. 1). The head and nape were shaved to minimize the loss of concussion impact. As soon as the rat showed a slight sign of waking up, the iron pendulum was released, and hit the head from a position 90 ~ to the vertical line of gravity. At the time of impact, the head moved slightly forward, and the pendulum swung further for a dis526
tance of 35 ~ from the vertical line. The lower end of the pendulum was marked with colored chalk and its point of impact could be identified in relation to the external occipital p r o t u b e r a n c e . R a t s not hit at the protuberance were not included in this report, and those that showed signs of hemorrhagic contusion were discarded. The control animals were left on the wooden board to wake up without being struck by the pendulum. The dynamic force of the impact was determined by trial and error in our pilot experiments. There was no loss of consciousness or any other evidence of concussion in the rats when hit by the pendulum dropped from a position 45 ~ to the vertical plane. When the pendulum was dropped from a horizontal position of 90 ~ to the line of gravity, the rats showed transitory unconsciousness for 3 to 10 minutes. Thus, the pendulum swung from 90 ~ to the vertical plane was selected for the experiments on 12 rats, and the energy absorbed by the head was calculated. The energy absorbed by the rat's head was determined from the difference in potential energy corresponding to the initial and final positions of the pendulum. If the initial angle and the angle after impact are Ao and At, respectively, the difference in potential energy ( E o - E~) can be calculated as: Eo - Er = W
R9 (cos At - cos Ao),
where W is the weight of the pendulum and R is the distance from the center of gravity to the support along the pendulum (Fig. 2). Friction energy loss o f the testing system was determined by measuring the difference between the initial and final angles in a "freeswing" situation. This angle due to friction loss was subtracted from the initial angle in computing the energy absorption of the animal. In the experiments conducted, the initial angle was always set at 90 ~. This had an angle of friction loss of 5 ~ in the system. Therefore an initial angle of 85 ~ was used. Other values used in the computation were: W = 1505 gm and R = 13.1 cm. The energy absorption curve is given in Fig. 2. The dotted line shows that for a 35 ~ angle after impact, the energy absorbed is approximately 1450 g m / c m . The 12 rats were killed at the following times after concussion: two at 30 minutes, four at 1 hour, two at 2 hours, two at 6 hours, J. Neurosurg. / Volume 47 / October, 1977
Experimental cerebral concussion and two at 24 hours. The rats were anesthetized by an intraperitoneal injection of a 3.5% chloral hydrate solution (1 ml/100 gm body weight). Cadmium-free ferritin in the amount of 0.3 to 0.5 ml (4.5 mg/ml) was injected into the tail vein, 5 to 15 minutes before death. The animals were killed by perfusion of 20 to 30 ml of warm phosphate buffer (0.1 M at 37 ~ C) through the ascending aorta, followed by 100-to 120 ml of cold 4% glutaraldehyde (in 0. I M phosphate buffer at 4 ~ C). Immediately after the perfusionfixation the cranium was opened, and the brain and upper cervical spinal cord removed. The following specimens were collected: the right and left frontal cortex and subcortical white matter, the right and left occipital cortex and subcortical white matter, the cerebellar cortex and white matter, both superior colliculi, the tegmentum of the medulla oblongata, and the gray and white matter of the first two segments of the cer,r cord. All specimens were postfixed in cold osmium tetroxide for 90 minutes, dehydrated in graded ethanols, and embedded in Maraglas mixture. Thick (1 um) and thin (60 nm) sections were cut from each specimen on a PorterBlum MT-2 ultramicrotome.* The thick sections were stained with 0.1% toluidine blue for light microscopy. The thin sections were stained with uranyl acetate and lead citrate for examination by a JEOLCO 100B electron microscope.t Results
All the rats were unconscious for 3 to 10 minutes after the concussion. During this time the corneal and light reflexes were absent and the animals did not respond to painful stimulation. After they awakened, they remained sluggish for 15 to 20 minutes. Thereafter they behaved and fed normally, and showed no evidence of focal or generalized neurological abnormality. On gross examination the brains of the concussed rats were normal. In particular there was no evidence of swelling or hemorrhages. *Porter-Blum MT-2 ultramicrotome manufactured by Ivan Sorvall, Inc., Norwalk, Connecticut. tElectron microscope supplied by Jeol, Inc., 477 Riverside Avenue, Medford, Massachusetts. J. Neurosurg. / Volume 4 7 / October, 1977
t
!o o
90*~
90*
,;,' -%.+
k':'G:iV'-.--__i_~"~"W*~
//
_ t-inal ~//
i
~t
DATUM
/
~ot.6 ~t.4 21.2 g
0.1 5~
O~
35"
40 ~
45"
50 ~
55 ~
Angle offer impact
FIG. 2. Upper."Diagram showing the initial and final positions of the iron rod with reference to the line of gravity where the rat's head received the impact. Lower."Graph representing the variation in energy absorption as a function of the angle after impact, for an initial angle of 90*. For a postimpact angle of 35 ~ the energy absorbed is 1450 gm/cm. The curve has been adjusted for friction energy loss of the testing system. Light microscopy revealed slight edema in the frontal white matter, the medulla, and upper cervical cord. A few neurons in the occipital cortex and a considerable number of neurons in the cervical cord appeared slightly swollen but, in contradiction to previous findings,* there was no evidence of chromatolysis in the nerve cells. In general, the central nervous system specimens showed only minor changes under the light microscope. Electron microscopy demonstrated significant changes in the brain ultrastructure of the concussed rats; the abnormalities showed great regional variation. 527
L. Bakay, J. C. Lee, G. C. Lee and J. R. Peng
FIG. 3. Left: Occipital cortex of control rat revealing intact neuronal mitochondria and narrow extracellular spaces. X 15,000. Right: Occipital cortex in an experimer]tal rat 1 hour after concussion. Many neuronal mitochondria (m) are moderately swollen, and their cristae are pushed to the periphery. The ribosomes (R), the endoplasmic reticulum (Er) (Nissl substance), the Golgi complexes (G), and the nucleus (N) are normal. X 5000.
The brains of control animals revealed a well known ultrastructural picture of intact mitochondria and narrow extracellular spaces (Fig. 3 left). By comparison, in the occipital cortex of the concussed rats, the point of impact, m a n y mitochondria of the neurons were swollen (Fig. 3 right). The mitochondrial cri~tae were pushed to the periphery by excess fluid. T h e m i t o c h o n d r i a of glial cells remained normal. The nucleus, Nissl substance, and Golgi complex of the neurons were not altered, nor was there any evidence of chromatolysis. There was no edema in the occipital cortex or white matter. Changes in the cerebellar cortex were similar to those found in the occipital lobe. The superior colliculi remained entirely intact. The most severe damage occurred in the medulla oblongata and the upper cervical cord. Here numerous neuronal mitochondria were found to be greatly swollen (Fig. 4 left). In some instances the cristae could barely be 528
seen because they were pushed against the outer membrane. There was slight enlargement of the extracellular space in the neuropil. The white matter, on the other hand, showed considerable extracellular edema (Fig. 4 right). The axons revealed slight overhydration. There was some ruffling of the myelin sheaths. A great concentration of gliofilaments was present in the processes of the fibrous astrocytes. The frontal cortex, the site of the contre coup, revealed minor changes in terms of slight swelling of the perivascular astrocytic processes and mild enlargement of the extracellular space (Fig. 5 left). Although the "tight junction" of the capillary endothelium was not widened, numerous ferritin particles were present in the basal lamina of the capillaries and venules. A few ferritin particles were even seen in the extracellular space, indicating increased permeability of the blood-brain barrier. There was occasional
J. Neurosurg. / Volume 47 / October, 1977
Experimental cerebral concussion
FIG. 4. Left: Neuropil of the gray matter of the upper cervical cord 1 hour after concussion. Numerous neuronal mitochondria (m) are swollen, and most of their cristae can be seen only over part of the periphery. Other subcellular organelles and synapses (s) appear intact. The extracellular space (e) is slightly enlarged. X 3000. Right: White matter of the upper cervical cord 1 hour after concussion. The extracellular space (e) is greatly enlarged. The myelin sheaths (Ms) of some nerve fibers show slight rufflings. The axons (Ax) appear somewhat overhydrated. A great concentration of gliofilaments (Gf) is evident in the processes of fibrous astrocytes. • 8500.
FIG. 5. Left: Frontal cortex 1 hour after concussion. The perivascular astrocytic processes (A) are moderately swollen. The extracellular space (e) in the neuropil is slightly enlarged. A few mitochondria (m) appear partially hydrated. The "tight junction" (T) of the endothelial cells is not widened. Numerous ferritin particles can be seen in the basal lamina (B) of a capillary, and some in the swollen astrocytic processes and the enlarged extracellular space (arrows). L = lumen of the capillary. X 4000. Right: Subcortical white matter of the frontal lobe 1 hour after concussion. The extracellular space (e) is enlarged. The axons (Ax) and their mitochondria (m) appear intact. The myelin sheaths of some nerve fibers show rufflings (Ms). • 3250.
J. Neurosurg. / Volume 4 7 / October, 1977
529
L. Bakay, J. C. L e e , G. C. L e e a n d J. R. P e n g mild swelling of the neuronal and glial mitochondria. In the frontal white matter, there was a definite extracellular edema as evidenced by the enlarged extracellular spaces (Fig. 5 right). Some ruffling of the laminae of the myelin sheaths was also observed. These alterations were already observed 30 minutes after the concussion but in a lesser degree of severity. They reached their peak at 1 hour, declined afterward, and virtually disappeared by 24 hours. The severe changes of the mitochondria were completely reversible within 24 hours, while a slight enlargement of the extracellular space of the white matter in the frontal lobes and upper cervical cord remained. The same changes were observed in the brains of rats sacrificed in the same period of time after concussion. Discussion
Previous studies of experimental brain concussion were performed on animals without a known force of the impact on the skull or with a calculated force but after exposure of the brain. 8,9 More accurate data were obtained from the experiments of Ommaya, et al. ~~ We believe that the simple adjustable pendulum used in our experiments, exerting its force on the intact skull, comes closest to the situation prevailing in the average head injury in man that results in cerebral concussion, because the transient unconsciousness is followed by complete recovery without any focal neurological signs. It is still amenable to the calculation of the force of impact in dynamic terms, readily providing us with the actual energy absorbed by the head at impact. Our observations confirm the statement made by Brodersen 3 that the severity of clinical symptoms is proportional to the energy absorbed by the head on impact. Our findings also indicate that the major damage occurs in the frontal region and at the craniospinal junction, while the mesencephalon and the caudal diencephalon are the least vulnerable. This is comparable with the findings by Ommaya and Gennarelli. 1~ From a mechanical point of view, the changes are caused by contre coup as far as the frontal lobes are concerned and stretchshearing effects at the craniospinal junction. In summarizing the effect of concussion at the ultrastructural level, mitochondrial changes occur in the neurons of the occipital 530
cortex, the region of the impact (Fig. 3 right). The changes in the structure of the neuronal mitochondria are even more severe in the medulla and upper cervical cord, areas somewhat remote from the impact but subject to traumatic stretch and shearing (Fig. 4). However, it should be pointed out that swelling of neuronal mitochondria is not specific to brain concussion. It was observed in a variety of neurological diseases including acute and chronic hypoxia and radiation damage, to name a f e w . 2,15,17 Interestingly enough, edema does not develop in the impact region. It occurs in areas farther away, such as the frontal lobes and the craniospinal junction. The fact that the mesencephalon, as represented by the superior colliculi, remains intact can be explained by the basic mechanical principles involved. The skull and its contents can be viewed as an oval globe, round in cross section and oval in horizontal section. Any force hitting the skull at one point can be resolved in two component forces that meet at the farthest opposite point, giving rise to a contre-coup effect and to a shearing action at the craniospinal junction, s Since the superior colliculi are close to the center of globe, they receive zero force from any impact to the periphery of the ovoid mass. The localized cerebral edema caused by concussion is of the vasogenic type and reflects the transient increase in the permeability of the blood-brain barrier as shown by the extravasation of ferritin. Similar findings were reported earlier with the use of methylene blue, TM fluorescein, 1~ Evans blue-albumin, 1~ and W2.4 In most experiments utilizing vital dyes, increased permeability was restricted to the first 24 hours. Previous experiments by Bakay, et al., 1 using nuclear magnetic resonance technique revealed that the edema fluid in trauma consists of two components. One is motionally restricted, " b o u n d " water, the other motionally unrestricted, "bulk" water. The rapid development, shift, and disappearance of excess water in concussion must be attributed to the fast moving water component. The sudden onset of mitochondrial swelling and its prompt disappearance in 24 hours is probably not caused by a complete disappearance and reappearance of mitochondrial enzymes. A more likely explanaJ. Neurosurg. / Volume 47 / October, 1977
Experimental cerebral concussion tion is a sudden increase in the permeability of the mitochondrial outer m e m b r a n e . The inward flow of excess fluid results in a swelling of the organelle, which pushes the cristae to the p e r i p h e r y and dilutes the m i t o chondrial enzymes. H a m b e r g e r and Rinder 9 found a brief increase in the succinoxidase activity and concluded that concussion activated the existent enzyme in the neuronal mitochondria rather than increasing the synthesis of the enzyme. Detailed histochemical studies now in progress in our laboratory might shed light on this aspect of cerebral concussion. These morphological investigations do not by t h e m s e l v e s explain the loss of consciousness in concussion. H o w e v e r , the edematous and mitochondrial changes of the medulla and upper cervical cord m a y explain such clinical findings as respiratory distress, cardiovascular alterations, and slowing of cerebral circulation? ,7,x~ References
1. Bakay L, Kurland R J, Parrish RG, et al: Nuclear magnetic resonance studies in normal and edematous brain tissue. Exp Brain Res 23:241-248, 1975 2. Bakay L, Lee JC: The effect of acute hypoxia and hypercapnia on the ultrastructure of the central nervous system. Brain 91:697-706, 1968 3. Brodersen P: Posthypoxic and traumatic coma: aspects of coma mechanisms, in Ingvar DH, Lassen NA (eds): Brain Work. The Coupling of Function, Metabolism and Blood Flow in the Brain. Copenhagen: Munksgaard, 1975, pp 269-279 4. Cassen B, Neff R: Blood-brain barrier behavior during temporary concussion. Am J Physiol 198:1296-1298, 1960 5. Denny-Brown D: Brain trauma and concussion. Arch Neurnl 5:1-3, 1961 6. Denny-Brown D, Russell WR: Experimental cerebral concussion. Brain 64:93-164, 1941 7. Gronwall D, Wrightson P: Cumulative effect of concussion. Lancet 2:995-997, 1975 8. Gurdjian ES: Recent advances in the study of the mechanism of impact injury of the head - A summary. Clin Neurosurg 19:1-42, 1972
J. Neurosurg. / Volume 47 / October, 1977
9. Hamberger A, Rinder L: Experimental brain concussion. J Neuropathol Exp Neurol 25:68-75, 1966 10. Ommaya AK, Gennarelli TA: Cerebral concussion and traumatic unconsciousness. Correlation of experimental and clinical observations on blunt head injuries. Brain 97:633-654, 1974 11. Ommaya AK, Rockoff SD, Baldwin M, et al" Experimental concussion. A first report. J Neurosurg 21:249-264, 1964 12. Quadbeck G, Randerath K: Enfluss der Gehirnerschfitterung auf den Obertritt yon Methylenblau aus dem Blut in das Gehirn der Katze. Z Naturforsch 10B:168-173, 1955 13. Rinder L, Olsson Y: Studies on vascular permeability changes in experimental brain concussion. I. Distribution of circulating fluorescent indicators in brain and cervical cord after sudden mechanical loading of the brain. Acta Neuropathoi 11:183-200, 1968 14. Rinder L, Olsson Y: Studies on vascular permeability changes in experimental brain concussion. II. Duration of altered permeability. Acta Neuropathol 11:201-209, 1968 15. Roizin L, Orlovskaja D, Liu JC, et al: Acid-/3glycerophosphatase reaction products in the central nervous system mitochondria following x-ray irradiation. J Histocbem Cytocbem 23:402-410, 1975 16. Ward AA Jr: The physiology of concussion. Clin Neurosurg 12:95-111, 1966 17. Yu MC, Bakay L, Lee JC: Ultrastructure of the central nervous system after prolonged hypoxia. I. Neuronal alterations. Acta Neuropathol 22:222-234, 1972
This paper was presented at the Annual Meeting of the American Association of Neurological Surgeons in Toronto, Canada, April 24-28, 1977. This work was supported by the Buswell Foundation and, in part, by the Research Institute of Alcoholism, Buffalo, New York. Address reprint requests to: Louis Bakay, M.D., Department of Neurosurgery, Clinical Center, 462 Grider Street, Buffalo, New York 14215.
531
