- Email: [email protected]
Intracranial pressure during epileptic seizures
Electroencephalography and clinical Neurophysiology, 1984, 5 7 : 4 9 7 - 5 0 6
497
Elsevier Scientific Publishers Ireland, Ltd.
Clinical Section INTRACRANIAL PRESSURE DURING EPILEPTIC SEIZURES A N D R E W J. G A B O R *, A L L E N G. BROOKS *, ROBI:RT P. SCOBEY * and GIBBE H. PARSONS **
• Department of Neurology, and ** Department of Medi¢;nt,, Universi(v of California, Davis Medical Center, Sacramento, CA 95817 (U.S.A.) (Accepted for publication: January 13, 1984)
An epileptic seizure may, on occasion, be associated with a measurable transient elevation of intracranial pressure (ICP) that is temporally related to the electrographic and clinical manifestations of the seizure. Several reports have documented this association which appears to be independent of the type of seizure (partial, generalized, convulsive, or non-convulsive) and seizure etiology. Minns and Brown (1978) documented elevated ICP in 5 patients during seizures. Two patients had generalized convulsive seizures, 1 patient had absence seizures, 1 patient had adversive seizures and 1 patient had myoclonic seizures. All of the patients were children, none were paralyzed. Blood pressure, heart rate and respirations were not measured. These authors argued that metabolic acidosis, hypoxemia, CO 2 retention, compensatory vasodilation and increased cerebral blood flow were the factors responsible for elevation of ICP in patients with seizures. White et al. (1961) documented elevations of ICP during seizures induced by pentylenetetrazol in subjects paralyzed with succinyl choline. In most cases elevations of the patients' blood pressure and heart rate preceded measured elevations of ICP. Temporal relationships were not determined; the measurements of pressure were random rather than continuous. Increased ICP has also been documented in patients with posttraumatic electrographic seizures. The elevated pressures occurred during seizures without clinical manifestations or while tile patients were paralyzed and mechanically ventilated (Marienne et al. 1979; Tsementzis et al. 1979). For the most part, previous reports have at-
tempted to relate the elevation of ICP to changes in cerebral blood flow (CBF). An increase in CBF with epileptic seizures is now well established. Ingvar (1973) studied 23 patients with epilepsy 'mainly of the focal cortical type' during the ictal state and in one case increased CBF was accompanied by an elevation of ICP. In contradistinction, blood flow in the region of the seizure focus was below normal during the interictal state. Increased cerebral blood flow occurs also in patients who are anesthetized and paralyzed during seizures induced by electroshock therapy (Brodersen et al. 1973). The type of seizure appears irrelevant. Increased blood flow has been reportedly associated with partial complex seizures, partial simple seizures, generalized non-convulsive and generalized convulsive seizures (Sakai et al. 1978). In experimental generalized seizures produced by pentylenetetrazol in monkeys, CBF increased by a mean of 264% (Plum et al. 1968). The temporal relationship between seizures, cerebral blood flow and ICP therefore appears established. Despite the quantitative data describing CBF and ICP, the relationship between these measures and the electrographic seizure remains qualitative and undefined. It is unknown, for example, how the duration of the seizure or the intensity of the seizure (as perhaps manifest by spike frequency) affects CBF and ICP. It is unknown whether EEG spikes (a manifestation of excitation) have an effect on CBF and ICP that is distinct from that of slow waves of the spike and wave complex (a presumed manifestation of physiologic inhibition). The current study demonstrates a quantitative relationship between the electrographic manifestations of seizures and intracranial
0013-4649/84/$03.00 © 1984 Elsevier Scientific Publishers Ireland, Ltd.
498
pressure. Data for this study were obtained from a paralyzed patient with recurrent partial and secondarily generalized seizures.
Methods and Material
Intracranial pressure was monitored by a Richmond Bolt and a digital display unit. The digital display of the ICP was manually transcribed onto the ongoing EEG record every time the display changed or at least every 60 sec. The continuously monitored ICP in mm Hg could then be correlated with the ongoing electrographic activity with an accuracy of + 1.0 mm Hg and _+1.0 see.
The EEG was recorded with an electroencephalograph using gold-plated disc electrodes. Electrode placement approximated the 10-20 international system and was modified because of pressure monitoring equipment on the head. For data analysis, one channel of the EEG was selected as best demonstrating the electrographic seizure activity. Spikes were counted over 5 sec intervals and recorded as the average number of spikes/sec over that interval. Blood pressure and heart rate were monitored by an arterial line and EKG. Any changes in pressure or heart rate were recorded manually on the EEG chart. Respirations were controlled by a respirator throughout the recording. Case report The patient was a 31-year-old male originally admitted to a community hospital with fever, chills, headaches, vomiting and diarrhea of 4 days duration. At admission, the patient was having recurrent right partial motor seizures with secondary generalization. Six days later he was transferred to the medical intensive care unit of the University of California, Davis Medical Center. On admission, he had a temperature of 39.4°C and was being mechanically ventilated. General physical examination was normal. He was unresponsive to commands. Cranial nerve function was normal. He moved his left upper extremities spontaneously and moved his left lower extremity in response to
A.J. GABOR ET AL.
painful stimulation. Clonic seizures of the face and right upper extremity occurred intermittently. A Babinski response was present on the left. Blood count, serum electrolytes, liver function tests and blood gases were normal. A lumbar puncture showed an opening pressure of 250 mm H 2 0 and examination of the fluid showed 161 red blood cells/mm 3, 12 monocytes/mm 3, a glucose of 86 mg% and a protein of 48 mg%. Cultures were negative. A CT scan showed only small subarachnoid cisterns. An initial EEG had slow, disorganized background activity and periodic epileptiform complexes occurred maximally in the left hemisphere (PLEDS). Inter-spike intervals varied from 1 to 3 sec and on occasion myoclonic movements accompanied the epileptiform complexes. During the recording, 3 organized ictal episodes were manifest by rhythmic spikes that were maximal in the left fronto-central region and were accompanied by a tonic-clonic sequence involving the right face and arm. A nuclide brain scan showed marked increased blood flow to the left hemisphere. Static images showed minimal increased activity in the central deep portions of the left hemisphere. A provisional diagnosis of herpes simplex encephalitis resulted in the initiation of adenine arabenoside therapy. A Richmond bolt was inserted to monitor intracranial pressure. A left temporal lobe biopsy showed no signs of inflammation and fluorescent antibody staining of the biopsy material was negative for herpes simplex. The seizures were eventually controlled with high doses of phenobarbital. The patient remained comatose, developed pneumonia, disseminated intravascular coagulation, hypotension, sepsis and renal failure. He expired 10 days after admission. At autopsy, hypoxic changes and mild flattening of the gyri were noted but there was no evidence of encephalitis. During the first few days of hospitalization at the University Medical Center, the patient had frequent partial motor seizures involving the right upper extremity and face. He was paralyzed with pancuronium in an attempt to control the elevation of intracranial pressure that accompanied each
INTRACRANIAL PRESSURE D U R I N G SEIZURES
seizure. Despite paralysis, intermittent elevations of intracranial pressure continued to occur and EEG monitoring was initiated. At the beginning of the recording, continuous seizure activity was present and persisted for 23 min. During this time, the intracranial pressure fluctuated between 17 and 30 mm of mercur~ (230-405 mm H20 ). Administration of 1 g of intravenous phenobarbital terminated the seizure and the intracranial pressure gradually fell to lC* mm of mercury (135 mm H20 ). Five relatively brief electrographic seizures were then recorded while the patient was paralyzed.
Results
Electrographic characteristics of the seizure The onset of each seizure was characterized by the appearance of moderate to high voltage
A
B
499
rhythmic spikes which were of maximal voltage in the left frontal and inferior frontal head regions. During the first 5-10 sec of each seizure, the spikes were well restricted to the left hemisphere. Subsequently, however, they occurred bilaterally and synchronously. The absolute voltage of the spikes was not determined since referential electrode montages were not used during the recording. The voltage differences between adjacent electrodes even at the onset of the seizure were 100-150/~V in magnitude. Fig. 1 is an example of the electrographic changes that occurred during one of the patient's seizures. At the onset of the seizure, rhythmic spikes occurred in the left frontal and inferior frontal regions. Within 5-10 sec, the seizure activity became generalized. The spike frequency changed during each seizure and the sequence of changes noted in Fig. 1 is typical of the patient's other seizures. Here, the average spike frequency
D
C
E
C3 - C Z cz-c4
?,s I
F'r-r3
I
SEIZURE DURATION 5
(SECONOS)
I CP (mmH9)
IO
35
55
70
TERMINATION
12.3
15
16.2
15.4
Fig. 1. A-E: examples of electrographic seizure activity. A - D are examples of seizure activity recorded at 5, 35, 55 and 70 sec after the onset of the seizure. The abrupt termination of the seizure occurs in E. In Figs. 2, 3, 4 and 6, this seizure activity is represented by the symbol ' © ' . The change in spike frequency during the course of the seizure is characteristic of all seizures. Cal., 100 ~V, 1 sec.
500
A.J. GABOR ET AL.
at the beginning of the seizure was 8.2 spikes/sec. The maximal spike frequency (14.3 spikes/sec) occurred several seconds later. A reduction of the spike frequency occurred before termination of the seizure. The change in spike frequency was characterized by two changes in wave form morphology. First, there was a change in the time course of the individual spikes. The duration of each spike at the base was longer at the lower frequencies and shorter at the higher frequencies. Second, near the end of each seizure, not only did the duralion of the spikes increase but also the spikes were interspersed with slow waves that were manifest as spike and wave complexes. F o u r of the seizures had similar durations (85 _ 5.3 sec). One of the seizures was shorter than the rest (48 sec, Fig. 2). For the 4 seizures of similar duration, the peak spike frequency occurred with a mean latency of 42 + 9.5 sec after seizure onset; the latency of the peak spike frequency for the briefer seizure was 15 sec. Each seizure terminated abruptly with a single spike, a spike and wave complex or a slow wave. The background activity was attenuated for approximately 5 - 1 0 sec following each seizure.
In the intervals between the organized electrographic seizure activity, isolated spikes, sharp waves, and polyspike and wave complexes occurred bilaterally and synchronously. They were generally of maximal voltage in the anterior quadrant on the left. At times, isolated wave forms were well restricted to this area (rather than appearing bilaterally) and increased in frequency of occurrence prior to the seizures. The blood pressure was monitored during each seizure. Interictally, the blood pressure varied between 98-115 m m H g systolic and 53-61 m m Hg diastolic pressure. During one seizure, the blood pressure increased to 1 1 6 / 6 7 m m Hg. In no case, however, was there a significant change in the blood pressure during the seizure that could be construed as being associated with the electrographic seizures themselves. In no instance, could the time course of any blood pressure change be correlated with changes in ICP. The patient's pulse rate also showed no consistent changes relative to the seizures. The recorded changes of ICP could not be attributed, therefore, to any secondary effects of the seizures such as changes in respiration, m o t o r activity, blood pressure or heart rate.
Relationship of electrographic seizure activity to changes in ICP 16
The increase in pressure above pre-ictal values during each of the 5 seizures was similar (6.8 + 0.9
14 12 Sp/$ 10
187
14" fit &
°P,
•
o'f
ooae~
..
•
,~ 12"
°;,¢
IO" ~=
• •
•= A
-2'0
~
z'o
,b
~
do
,~o
,~o
,~o
,~o
~o
2~o
SEC.
6-
~ Onset of Epileptocjenl¢Activity
-,'o ~
~
,~o
,~o
z~o
TqME (percent of seizure duration)
Fig. 2. Time course of ICP and spike frequency. The recorded
values for each seizure are plotted as a function of t~me. Ap (ram Hg) represents the difference between the recorded pressure during the seizure and the pressure recorded pri~r to the seizure. Time 0 represents the beginning of the seizure. Spike frequency is measured in spikes/see (Sp/s) and is drawn on the same time base as the ICP.
Fig. 3. The recorded ICPs are plotted for each of 5 seizures as a function of time. All graphs are normalized to seizure duration in order to demonstrate the similarities of ICP for each seizure as a function of seizure duration. Much of the difference in time course of the ICP can be accounted for by seizure duration and spike frequency.
INTRACRANIAL PRESSURE D U R I N G SEIZURES
mm Hg) (Fig. 2). The ICP prior to each seizure ranged between 8 and 10 mm of mercury (9.5 + 0.92). Therefore, the differences in the absolute magnitude of the peak ICP associated with the 5 seizures could be accounted for by the ICP pressure existing at the onset of each of the seizures (Fig. 3). After the onset of the electrographic seizure activity, the ICP increased slowly to a maximum During 4 seizures, peak pressure occurred after a mean interval of 70 + 6.7 sec after seizure onset The peak pressure of the brief seizure was recorded 28 sec after seizure onset. The time from seizure onset to peak pressure was almost twice that of seizure onset to peak spike frequency. Each seizure was characterized by a consistent temporal relationship between the peak ICP and the beginning and termination of the electrographic seizure: (1) peak pressure occurred after onset of the ictal epileptiform activity; (2) the ICP began to decline before termination of the electrographic seizure activity (Fig. 2). In addition, the time course of the ICP appeared to be related in some way to the organization of the electrical seizure activity, both during the ictal phase and during the recovery phase following the seizure (Fig. 2).
The recovery phase The recovery phase is defined as that interval of time beginning with the termination of the seizure and ending at the time the pressure returned to pre-ictal values (Fig. 3). During the recovery phase, the pressure change was non-linear. The pressure change tended to be more rapid when the pressure was high as compared to changes when the pressure was low. Although this non-linear change of pressure could fit many mathematical functions, an attempt was made to see how well the changes in pressure fit a simple exponential function as a first approximation. The choice of an exponential function in this context is reasonable since intracranial pressure-volume relationships are thought to be exponential within certain limits (Miller 1975; Sklar and Elashvili 1977; Marmarou et al. 1978; Tans et al. 1982). Since exponential functions are linear if plotted on semi-log paper, the log of the difference be-
501
81~
To=74,8.~
I
zxPs]
,-c
I
B1~
Tr==0. c=95.6 99
sec
I ;
'
7'0 SECONDS
Fig. 4. Semi-log plot of Ap (see legend Fig. 2) as a function of time during the recovery phase (immediate post-ictal period). Time 0 = time of termination of ictal activity. A linear regression function was plotted. The time constant (Tc) and regression coefficient (r 2) are indicated for each function. The high correlation coefficient indicates that the function may be considered exponential as a first approximation.
tween pre-ictal pressure and pressure during the recovery phase (Ap) was plotted as a function of time (Fig. 4). A straight line drawn through the data points describes the data well (correlation coefficients (r 2) above 0.90). The slope of this fine may be defined as the time required to decrease the pressure by a factor of 10. This duration may be termed the pressure-time index (see also Marmarou et al. 1978, for pressure-volume index). The time constant, which is the time required for
502
A.J. GABOR ET AL.
the pressure to d r o p to 37% of its original value, was calculated. T h e time c o n s t a n t was calculated as the reciprocal of the slope of the straight line r e p r e s e n t i n g the n a t u r a l log of the pressure difference with respect to time. T h e m e a n time constant for four of the seizures was 85.5 _+ 8.73 sec. T h e time c o n s t a n t for the brief seizure was 112.4 sec.
A.
EEG
B. SP,KES IIIIIIIIIIIIIIIIIIII
I
I
C. [-
I
"I
Seizure
I
u
Sec.
Fig. 5. Diagrammatic representation of seizure-ICP relationship. A, B: a spike during the seizure is considered a unitary event in time. C: bar graph depicting number of spikes occurring at 5 sec intervals. D: a constant increment of ICP with each spike is postulated. Following each increment ICP, the ICP declines exponentially toward prespike levels. The next spike increments the ICP from the residual value (o) existing at the time the spike occurred. For convenience of calculation, n spikes in a 5 sec period was considered to be an increment of ICP n times the value of a unitary ICP increment. For each bin of the histogram in C, a scaled ICP pulse is added to the residual pressure and the sum yields a function predicting ICP as a function of time. The predicted ICP function (A p = P~F', e -(T-xn)/Tc) is plotted. Ps = a scaling function (mm Hg/spike); Fn = average number of spikes/sec for 5 sec interval; T,: = time constant of the recovery phase; ( T - T n ) = an interval of time, referenced to the beginning of the exponential weighting function (Bower and Schultheiss 1958). A, B, C, and D are drawn on the same time base.
Ictal phase In c o n t r a d i s t i n c t i o n to the recovery phase, the changes of pressure as a function of time d u r i n g the ictal phase, d i d not fit a simple e x p o n e n t i a l function. To account for the time course of the I C P d u r i n g the ictal phase, the electrical events d u r i n g the seizure were studied. The f u n d a m e n t a l unit of ictal activity available for m e a s u r e m e n t is the E E G spike. The voltage of the spikes was relatively c o n s t a n t t h r o u g h o u t the ictus. F o r p u r p o s e s of this study, the a s s u m p t i o n was m a d e that each spike represents an i n d e p e n d e n t unit of p a t h o p h y s i o l o g i c seizure activity that i n c r e m e n t s the ICP. Viewed from this perspective, the pressure change d u r i n g the ictal phase could be due to short d u r a t i o n i n c r e m e n t s of I C P occurring with each spike a n d s u m m a t i n g over time (Fig. 5). E a c h i n c r e m e n t of I C P would then decrease tow a r d pre-ictal levels with a time course d e s c r i b e d for the recovery p h a s e until the occurrence of the next spike and pressure increment. C o m p l e t e p r e d i c t e d pressure curves could be d r a w n by s u m m a t i n g i n c r e m e n t s of I C P for every spike during the seizure (Fig. 6). The c o n t i n u o u s function thus g e n e r a t e d was scaled b y visual app r o x i m a t i o n to best fit the m e a s u r e d e x p e r i m e n t a l data. This scaling d i m e n s i o n e d each i n c r e m e n t of I C P in m m H g / s p i k e . The relationship of the p r e d i c t e d pressure curves to m e a s u r e d values of I C P are shown in Fig. 6 for all 5 seizures. It should be n o t e d that the m e a s u r e d values of I C P vary from the p r e d i c t e d values b y no m o r e than 2 m m Hg. This difference is within the range of error a t t r i b u t a b l e to the technique of m e a s u r e m e n t a n d r e c o r d i n g of data. In each case, the m a t h e m a t i c a l function closely a p p r o x i m a t e s the rate of increase of intracranial pressure, the p e a k i n t r a c r a n i a l pressure, the phase difference between m a x i m a l spike frequency a n d m a x i m a l ICP, and the rate at which the I C P returns to pre-ictal values after t e r m i n a t i o n of the seizures.
Discussion
The t e m p o r a l association between seizures a n d increased i n t r a c r a n i a l pressure is well d o c u m e n t e d in the literature. Previous w o r k r e p o r t e d on the
I N T R A C R A N I A L PRESSURE D U R I N G SEIZURES
503
A
B
i4 Sp/s to 6
Sp/s i
AP ij 0J • -~
6
,6o
-:;o 6
~6o
,bo
'
Sec.
'
26o
'
a~
Sec.
D
Sp/sto
SP/Sl2 AP4
AP4
2 -2b 6
too '
~ -2o
~ 0
;
!
...............
Sec.
,6o Sec.
LB~
F"
,o]
:t
'
{IP41 J
•
2
•
o] i"
-20
'1
I
0
" !
I00
|
/
Z00
See. Fig. 6. The ICP associated with the patient's 5 seizures are plotted as a function of time. Measured ICP is indicated by symbols ©, e, r,, A II. The calculated ICP function is represented by solid lines. Scaling functions (P~, Fig. 5) have an average value of 0.07 + 0.01 m m Hg/spike,
mechanisms directly responsible for the increased pressure such as documenting causes for increased cerebral blood flow. None have attempted to quantitate the transduction of the physio-pathologic event (the epileptogenic event) to changes in
ICP. The present study describes a model for extrapolating electrical seizure activity to ICP changes. The model accounts for many of the empirically observed characteristics of the relationship between seizure activity and ICP. These
504
include the following: (1) Seizure onset before onset of increased ICP. (2) Rate of increase in ICP. (3) Temporal relationship between peak spike frequency and peak ICP. (4) Temporal relationship between peak ICP and seizure termination. (5) Time course of ICP following termination of the seizure. Perhaps the most striking aspect of this study is that the time course of the ICP changes can be predicted on the basis of only two measurable values (the electrical spike frequency of the ictal phase and the time course of the ICP recovery phase). The time course of the recovery phase ~s presumed to represent a pressure response of the intracranial physical system to small changes in intracranial volume. The return of pressure to pre-ictal values is assumed to be due to normal CSF pressure-volume dynamics, i.e., it is a passive response to a previous increase in CSF volume. The time course of the ICP during the recovery phase is primarily dependent upon outflo~ resistance and the compartmentalization of ttae increased intracranial volume. Outflow resistance may be elevated due to restrictions of CSF flow or jugular/venous sinus obstruction. Transfer of fluid to the intracerebral compartment (edema) from the vascular compartment or CSF will also alter the time course of ICP during the recovery" phase. The presence of edema may be considered a form of increased outflow resistance. It is presumed that the pressure-time relationship of the recovery phase (Fig. 4) is related to the pressure-volume relationships described in other experimental studies (Marmarou et al. 1978: Sklar and Elashvili 1977). The change of ICP undoubtedly reflects changes in intracranial xolume originating in the vascular system. The distribution of the fluid within the cranium remains uncertain. In the case described here, there was a rapid return of the pressure to pre-ictal values. This suggests that most of the increased volume was within the intravascular compartment (vasodilation) rather than being distributed in the extravascular space (edema). If a large proportion of fluid were extravascular, such as might occur with prolonged
A.J. GABOR ET AL.
seizures, the cerebral compliance could decrease for extended periods of time. This in turn would increase the potential for greater peak pressures with each subsequent seizure. The data provided here show that the pressure-time function is not the same for the ictal and recovery phase. The time course of the change in pressure during the ictal phase was quantitatively related to the ongoing epileptogenic electrical activity. It may be argued that each ictal event (represented by the EEG spike) results in an average increment of intracranial pressure. This average increment of pressure will be dependent on multiple factors and is likely to vary between patients and between seizures in the same patient. Following each increment of ICP, the ICP declines toward pre-ictal levels. Each pressure increment adds to the residual pressure existing at the time the spike occurred. The model incorporates these concepts and predicts the following relationships between electrographic events and ICP: (1) The rate of pressure change will be variable during the ictal phase. (2) The rate of pressure change will be proportional to the spike frequency over short intervals of time (0.1 time constant). (3) The peak pressure will occur and be maintained when the spike frequency is sufficiently high to counteract the decline of ICP toward preictal levels. (4) The peak pressure will occur after the peak spike frequency with a time delay determined by the spike frequency and the time course of the recovery phase. (5) If the seizure continues beyond the time of the peak spike frequency, with spikes occurring at a lower rate, the pressure will decrease despite continuation of the ictal activity. The predicted values of ICP closely approximate the experimentally derived data. Therefore, the time course of the ICP appears to be determined by the summations of the fundamental units of abnormal synchronized activity (the epileptogenic spike) and the CSF pressure-volume dynamics existing at the time of the seizure. An average increment of ICP per spike can be calculated for each seizure.
INTRACRANIAL PRESSURE DURING SEIZURES It is presumed that most seizure activity will be accompanied by increased intracranial volume. This increased volume, however, may not be associated with an increased ICP if the system is highly compliant. A poorly compliant system will be associated with changes in pressure during a seizure and the lower the compliance the greater the pressure change. Therefore, it may be anticipated that some patients will show no evidence of change in 1CP during seizures. On the other hand, compliance will be reduced during prolonged seizures and result in much higher ICP than would be expected with brief seizures. This is exemplified by the ICP of the patient presented here following a prolonged seizure prior to anticonvulsant therapy. Using the model described here, the ICP after 10 min of continuous seizure activity can be predicted. Assuming an average spike frequency of 12 spikes/sec, a recovery phase time constant of 90 sec, and a pre-ictal pressure of 10 mm Hg, the ICP after 10 min of seizure activity would be 25 mm Hg or 338 mm H20. In the patient described here, pressures ranged from 230 mm H20 to 41)5 mm H20 during a continuous seizure of unknown duration. The risk of patients developing high ICP due to prolonged seizures will be significantly increased if patients are therapeutically paralyzed and the occurrence of seizure activity is difficult or impossible to recognize by clinical observation.
Summa~ A comatose 31-year-old male with presumed viral encephalitis and frequent partial motor seizures was paralyzed with pancuronium in an attempt to reduce recurrent elevation of intracranial pressure (ICP) associated with each seizure. ICP was continuously monitored with a Richmond Bolt and 5 electrographic seizures originating in the left frontal area were recorded. Each ictal episode was associated with stable blood pressure and an increase of ICP. The average seizure duration was 78 + 17 sec (mean :t S.D.) and the average maximum increase of ICP above baseline during the seizures was 6.5 + 0.6 mm Hg with average peak ICP of 16.0 + 0.86 mm Hg. A
505 simple mathematical model predicts the rate of increase of ICP, the peak ICP, the phase difference between maximum spike frequency and maximum ICP, and the rate at which ICP returns to pre-ictal values after termination of the seizure. The predicted values of ICP closely approximate the experimentally derived data. Therefore, the time course of the ICP appears to be determined by the frequency of the fundamental units of abnormal synchronized activity (the epileptigenic spike) and the CSF pressure-volume dynamics existing at the time of the seizure. An average increment of ICP per spike can be calculated for each seizure. The model also predicts that patients may develop high ICPs due to prolonged seizures. Prolonged unrecognized seizures may occur in patients who are therapeutically paralyzed as demonstrated by the case described here.
R6sum6 Pression intracrgmienne au cours de crises bpileptiques
Un sujet male de 31 ans dans un 6tat comateux pr6sentant une enc6phalite virale pr6sum6e et de fr6quentes crises 6pileptiques motrices partielles a 6t6 paralys6 au pancuronium afin de tenter de diminuer l'616vation r6p6t4e de la pression intracr~nienne (ICP) associ6e h chaque crise. L'ICP 6tait sous surveillance continue h l'aide d'un Richmond Bolt et 5 crises 61ectrographiques ayant leur origine dans l'aire frontale gauche ont 6t6 enregistr6es. Chaque 6pisode ictal 6tait accompagn6 d'une pression art6rielle stable et d'une augmentation de I'ICP. La dur6e moyenne de la crise a 6t6 de 7 8 + 17 sec (moyenne+ E.T.) et la moyenne de l'augmentation maximale de I'ICP au-dessus de la ligne de base au cours des crises, de 6,5 + 0,6 mm Hg avec un pic moyen de I'ICP h 16,0 + 0,86 mm Hg. Un mod61e math6matique simple permet de pr6dire le taux d'augmentation de I'ICP, le pic ICP, la diff6rence de phase entre la fr6quence maximum de pointes et le maximum de I'ICP, et la vitesse h laquelle I'ICP revient h des valeurs pr6ictales apr6s la fin de la crise.
506
Les valeurs pr~dites pour I'ICP sont une approximation tr6s proche des valeurs obtenues exp6rimentalement. Le d6cours temporel de I'ICP apparait donc d6termin~ par la sommation des unit6s fondamentales de l'activit6 anormale synchronis6e (pointe 6pileptog/me) et la dynamique pression-volume du liquide c6phalorachidien existant au moment de la crise. Une valeur movenne de l'augmentation de I'ICP par pointe peut ~tre calcul6e pour chaque crise. Le mod61e pr6dit aussi que les patients peuvent d6velopper des ICP 61ev6es lors de crises prolong6es. Des crises prolong6es non reconnues peuvent survenir chez des patients qui sont paralys6s th6rapeutiquement comme le d6montre te cas d6crit ici. References Bower, J.L. and Schultheiss, P.M. Weighting functions. In: Introduction to the Design of Servomechanisms Wiley, New York, 1958: 48-50. Brodersen, P., Paulson, O.B., Bolwig, T.G., Rogon, Z.E., Rafaelsen, O.J. and Lassen, N.A. Cerebral hyperemia in electrically induced epileptic seizures. Arch. Neurol. (Chic.), 1973, 28: 334-338.
A.J. GABOR ET AL. lngvar, D.H. CBF and electrical activity. Stroke, 1973, 4: 359. Marienne, J.P., Robert, G. and Bagnat, E. Post-traumatic acute rise of ICP related to sub-clinical epileptic seizures. Acta neurochir. (Wien), 1979, 28 (Suppl.): 89-92. Marmarou, A., Shulman, K. and Rosende, R.M. A nonlinear analysis of the cerebrospinal fluid system and intracranial pressure dynamics. J. Neurosurg., 1978, 48: 332-344. Miller, J.D. Volume and pressure in cerebrospinal axis. Clin. Neurosurg., 1975, 22: 76-105. Minns, R.A. and Brown, J.K. Intracranial pressure changes associated with childhood seizures. Develop. Med. Child. Neurol., 1978, 20: 561-569. Plum, F., Posner, J.B. and Troy, B. Cerebral metabolic and circulatory responses to induced convulsions in animals. Arch. Neurol. (Chic.), 1968, 18: 1-13. Sakai, F , Meyer, S., Naritomi, H. and Hsu, M. Regional cerebral blood flow and EEG in patients with epilepsy. Arch. Neurol. (Chic.), 1978, 35: 648-657. Sklar, F.H. and ElashvilL I. The pressure-volume function of brain elasticity. J. Neurosurg., 1977, 47: 670-679. Tans, J.TJ. and Poortviet, D.CJ. Intracranial volume-pressure relationship in man. Part l: Calculation of pressure-volume index. J. Neurosurg., 1982, 56: 524-528. Tsementzis, S.A., Gillingham, FJ. and Hitchcock, E.R. The effect of focal twitching on the intracranial pressure during paralysis and mechanical ventilation. Ann. clin. Res., 1979, 11: 253-257. White, P.T., Grant, P., Mosier, J. and Craig, A. Changes in cerebral dynamics associated with seizures. Neurology (Minneap.), 1961, 11: 354-361.
497
Elsevier Scientific Publishers Ireland, Ltd.
Clinical Section INTRACRANIAL PRESSURE DURING EPILEPTIC SEIZURES A N D R E W J. G A B O R *, A L L E N G. BROOKS *, ROBI:RT P. SCOBEY * and GIBBE H. PARSONS **
• Department of Neurology, and ** Department of Medi¢;nt,, Universi(v of California, Davis Medical Center, Sacramento, CA 95817 (U.S.A.) (Accepted for publication: January 13, 1984)
An epileptic seizure may, on occasion, be associated with a measurable transient elevation of intracranial pressure (ICP) that is temporally related to the electrographic and clinical manifestations of the seizure. Several reports have documented this association which appears to be independent of the type of seizure (partial, generalized, convulsive, or non-convulsive) and seizure etiology. Minns and Brown (1978) documented elevated ICP in 5 patients during seizures. Two patients had generalized convulsive seizures, 1 patient had absence seizures, 1 patient had adversive seizures and 1 patient had myoclonic seizures. All of the patients were children, none were paralyzed. Blood pressure, heart rate and respirations were not measured. These authors argued that metabolic acidosis, hypoxemia, CO 2 retention, compensatory vasodilation and increased cerebral blood flow were the factors responsible for elevation of ICP in patients with seizures. White et al. (1961) documented elevations of ICP during seizures induced by pentylenetetrazol in subjects paralyzed with succinyl choline. In most cases elevations of the patients' blood pressure and heart rate preceded measured elevations of ICP. Temporal relationships were not determined; the measurements of pressure were random rather than continuous. Increased ICP has also been documented in patients with posttraumatic electrographic seizures. The elevated pressures occurred during seizures without clinical manifestations or while tile patients were paralyzed and mechanically ventilated (Marienne et al. 1979; Tsementzis et al. 1979). For the most part, previous reports have at-
tempted to relate the elevation of ICP to changes in cerebral blood flow (CBF). An increase in CBF with epileptic seizures is now well established. Ingvar (1973) studied 23 patients with epilepsy 'mainly of the focal cortical type' during the ictal state and in one case increased CBF was accompanied by an elevation of ICP. In contradistinction, blood flow in the region of the seizure focus was below normal during the interictal state. Increased cerebral blood flow occurs also in patients who are anesthetized and paralyzed during seizures induced by electroshock therapy (Brodersen et al. 1973). The type of seizure appears irrelevant. Increased blood flow has been reportedly associated with partial complex seizures, partial simple seizures, generalized non-convulsive and generalized convulsive seizures (Sakai et al. 1978). In experimental generalized seizures produced by pentylenetetrazol in monkeys, CBF increased by a mean of 264% (Plum et al. 1968). The temporal relationship between seizures, cerebral blood flow and ICP therefore appears established. Despite the quantitative data describing CBF and ICP, the relationship between these measures and the electrographic seizure remains qualitative and undefined. It is unknown, for example, how the duration of the seizure or the intensity of the seizure (as perhaps manifest by spike frequency) affects CBF and ICP. It is unknown whether EEG spikes (a manifestation of excitation) have an effect on CBF and ICP that is distinct from that of slow waves of the spike and wave complex (a presumed manifestation of physiologic inhibition). The current study demonstrates a quantitative relationship between the electrographic manifestations of seizures and intracranial
0013-4649/84/$03.00 © 1984 Elsevier Scientific Publishers Ireland, Ltd.
498
pressure. Data for this study were obtained from a paralyzed patient with recurrent partial and secondarily generalized seizures.
Methods and Material
Intracranial pressure was monitored by a Richmond Bolt and a digital display unit. The digital display of the ICP was manually transcribed onto the ongoing EEG record every time the display changed or at least every 60 sec. The continuously monitored ICP in mm Hg could then be correlated with the ongoing electrographic activity with an accuracy of + 1.0 mm Hg and _+1.0 see.
The EEG was recorded with an electroencephalograph using gold-plated disc electrodes. Electrode placement approximated the 10-20 international system and was modified because of pressure monitoring equipment on the head. For data analysis, one channel of the EEG was selected as best demonstrating the electrographic seizure activity. Spikes were counted over 5 sec intervals and recorded as the average number of spikes/sec over that interval. Blood pressure and heart rate were monitored by an arterial line and EKG. Any changes in pressure or heart rate were recorded manually on the EEG chart. Respirations were controlled by a respirator throughout the recording. Case report The patient was a 31-year-old male originally admitted to a community hospital with fever, chills, headaches, vomiting and diarrhea of 4 days duration. At admission, the patient was having recurrent right partial motor seizures with secondary generalization. Six days later he was transferred to the medical intensive care unit of the University of California, Davis Medical Center. On admission, he had a temperature of 39.4°C and was being mechanically ventilated. General physical examination was normal. He was unresponsive to commands. Cranial nerve function was normal. He moved his left upper extremities spontaneously and moved his left lower extremity in response to
A.J. GABOR ET AL.
painful stimulation. Clonic seizures of the face and right upper extremity occurred intermittently. A Babinski response was present on the left. Blood count, serum electrolytes, liver function tests and blood gases were normal. A lumbar puncture showed an opening pressure of 250 mm H 2 0 and examination of the fluid showed 161 red blood cells/mm 3, 12 monocytes/mm 3, a glucose of 86 mg% and a protein of 48 mg%. Cultures were negative. A CT scan showed only small subarachnoid cisterns. An initial EEG had slow, disorganized background activity and periodic epileptiform complexes occurred maximally in the left hemisphere (PLEDS). Inter-spike intervals varied from 1 to 3 sec and on occasion myoclonic movements accompanied the epileptiform complexes. During the recording, 3 organized ictal episodes were manifest by rhythmic spikes that were maximal in the left fronto-central region and were accompanied by a tonic-clonic sequence involving the right face and arm. A nuclide brain scan showed marked increased blood flow to the left hemisphere. Static images showed minimal increased activity in the central deep portions of the left hemisphere. A provisional diagnosis of herpes simplex encephalitis resulted in the initiation of adenine arabenoside therapy. A Richmond bolt was inserted to monitor intracranial pressure. A left temporal lobe biopsy showed no signs of inflammation and fluorescent antibody staining of the biopsy material was negative for herpes simplex. The seizures were eventually controlled with high doses of phenobarbital. The patient remained comatose, developed pneumonia, disseminated intravascular coagulation, hypotension, sepsis and renal failure. He expired 10 days after admission. At autopsy, hypoxic changes and mild flattening of the gyri were noted but there was no evidence of encephalitis. During the first few days of hospitalization at the University Medical Center, the patient had frequent partial motor seizures involving the right upper extremity and face. He was paralyzed with pancuronium in an attempt to control the elevation of intracranial pressure that accompanied each
INTRACRANIAL PRESSURE D U R I N G SEIZURES
seizure. Despite paralysis, intermittent elevations of intracranial pressure continued to occur and EEG monitoring was initiated. At the beginning of the recording, continuous seizure activity was present and persisted for 23 min. During this time, the intracranial pressure fluctuated between 17 and 30 mm of mercur~ (230-405 mm H20 ). Administration of 1 g of intravenous phenobarbital terminated the seizure and the intracranial pressure gradually fell to lC* mm of mercury (135 mm H20 ). Five relatively brief electrographic seizures were then recorded while the patient was paralyzed.
Results
Electrographic characteristics of the seizure The onset of each seizure was characterized by the appearance of moderate to high voltage
A
B
499
rhythmic spikes which were of maximal voltage in the left frontal and inferior frontal head regions. During the first 5-10 sec of each seizure, the spikes were well restricted to the left hemisphere. Subsequently, however, they occurred bilaterally and synchronously. The absolute voltage of the spikes was not determined since referential electrode montages were not used during the recording. The voltage differences between adjacent electrodes even at the onset of the seizure were 100-150/~V in magnitude. Fig. 1 is an example of the electrographic changes that occurred during one of the patient's seizures. At the onset of the seizure, rhythmic spikes occurred in the left frontal and inferior frontal regions. Within 5-10 sec, the seizure activity became generalized. The spike frequency changed during each seizure and the sequence of changes noted in Fig. 1 is typical of the patient's other seizures. Here, the average spike frequency
D
C
E
C3 - C Z cz-c4
?,s I
F'r-r3
I
SEIZURE DURATION 5
(SECONOS)
I CP (mmH9)
IO
35
55
70
TERMINATION
12.3
15
16.2
15.4
Fig. 1. A-E: examples of electrographic seizure activity. A - D are examples of seizure activity recorded at 5, 35, 55 and 70 sec after the onset of the seizure. The abrupt termination of the seizure occurs in E. In Figs. 2, 3, 4 and 6, this seizure activity is represented by the symbol ' © ' . The change in spike frequency during the course of the seizure is characteristic of all seizures. Cal., 100 ~V, 1 sec.
500
A.J. GABOR ET AL.
at the beginning of the seizure was 8.2 spikes/sec. The maximal spike frequency (14.3 spikes/sec) occurred several seconds later. A reduction of the spike frequency occurred before termination of the seizure. The change in spike frequency was characterized by two changes in wave form morphology. First, there was a change in the time course of the individual spikes. The duration of each spike at the base was longer at the lower frequencies and shorter at the higher frequencies. Second, near the end of each seizure, not only did the duralion of the spikes increase but also the spikes were interspersed with slow waves that were manifest as spike and wave complexes. F o u r of the seizures had similar durations (85 _ 5.3 sec). One of the seizures was shorter than the rest (48 sec, Fig. 2). For the 4 seizures of similar duration, the peak spike frequency occurred with a mean latency of 42 + 9.5 sec after seizure onset; the latency of the peak spike frequency for the briefer seizure was 15 sec. Each seizure terminated abruptly with a single spike, a spike and wave complex or a slow wave. The background activity was attenuated for approximately 5 - 1 0 sec following each seizure.
In the intervals between the organized electrographic seizure activity, isolated spikes, sharp waves, and polyspike and wave complexes occurred bilaterally and synchronously. They were generally of maximal voltage in the anterior quadrant on the left. At times, isolated wave forms were well restricted to this area (rather than appearing bilaterally) and increased in frequency of occurrence prior to the seizures. The blood pressure was monitored during each seizure. Interictally, the blood pressure varied between 98-115 m m H g systolic and 53-61 m m Hg diastolic pressure. During one seizure, the blood pressure increased to 1 1 6 / 6 7 m m Hg. In no case, however, was there a significant change in the blood pressure during the seizure that could be construed as being associated with the electrographic seizures themselves. In no instance, could the time course of any blood pressure change be correlated with changes in ICP. The patient's pulse rate also showed no consistent changes relative to the seizures. The recorded changes of ICP could not be attributed, therefore, to any secondary effects of the seizures such as changes in respiration, m o t o r activity, blood pressure or heart rate.
Relationship of electrographic seizure activity to changes in ICP 16
The increase in pressure above pre-ictal values during each of the 5 seizures was similar (6.8 + 0.9
14 12 Sp/$ 10
187
14" fit &
°P,
•
o'f
ooae~
..
•
,~ 12"
°;,¢
IO" ~=
• •
•= A
-2'0
~
z'o
,b
~
do
,~o
,~o
,~o
,~o
~o
2~o
SEC.
6-
~ Onset of Epileptocjenl¢Activity
-,'o ~
~
,~o
,~o
z~o
TqME (percent of seizure duration)
Fig. 2. Time course of ICP and spike frequency. The recorded
values for each seizure are plotted as a function of t~me. Ap (ram Hg) represents the difference between the recorded pressure during the seizure and the pressure recorded pri~r to the seizure. Time 0 represents the beginning of the seizure. Spike frequency is measured in spikes/see (Sp/s) and is drawn on the same time base as the ICP.
Fig. 3. The recorded ICPs are plotted for each of 5 seizures as a function of time. All graphs are normalized to seizure duration in order to demonstrate the similarities of ICP for each seizure as a function of seizure duration. Much of the difference in time course of the ICP can be accounted for by seizure duration and spike frequency.
INTRACRANIAL PRESSURE D U R I N G SEIZURES
mm Hg) (Fig. 2). The ICP prior to each seizure ranged between 8 and 10 mm of mercury (9.5 + 0.92). Therefore, the differences in the absolute magnitude of the peak ICP associated with the 5 seizures could be accounted for by the ICP pressure existing at the onset of each of the seizures (Fig. 3). After the onset of the electrographic seizure activity, the ICP increased slowly to a maximum During 4 seizures, peak pressure occurred after a mean interval of 70 + 6.7 sec after seizure onset The peak pressure of the brief seizure was recorded 28 sec after seizure onset. The time from seizure onset to peak pressure was almost twice that of seizure onset to peak spike frequency. Each seizure was characterized by a consistent temporal relationship between the peak ICP and the beginning and termination of the electrographic seizure: (1) peak pressure occurred after onset of the ictal epileptiform activity; (2) the ICP began to decline before termination of the electrographic seizure activity (Fig. 2). In addition, the time course of the ICP appeared to be related in some way to the organization of the electrical seizure activity, both during the ictal phase and during the recovery phase following the seizure (Fig. 2).
The recovery phase The recovery phase is defined as that interval of time beginning with the termination of the seizure and ending at the time the pressure returned to pre-ictal values (Fig. 3). During the recovery phase, the pressure change was non-linear. The pressure change tended to be more rapid when the pressure was high as compared to changes when the pressure was low. Although this non-linear change of pressure could fit many mathematical functions, an attempt was made to see how well the changes in pressure fit a simple exponential function as a first approximation. The choice of an exponential function in this context is reasonable since intracranial pressure-volume relationships are thought to be exponential within certain limits (Miller 1975; Sklar and Elashvili 1977; Marmarou et al. 1978; Tans et al. 1982). Since exponential functions are linear if plotted on semi-log paper, the log of the difference be-
501
81~
To=74,8.~
I
zxPs]
,-c
I
B1~
Tr==0. c=95.6 99
sec
I ;
'
7'0 SECONDS
Fig. 4. Semi-log plot of Ap (see legend Fig. 2) as a function of time during the recovery phase (immediate post-ictal period). Time 0 = time of termination of ictal activity. A linear regression function was plotted. The time constant (Tc) and regression coefficient (r 2) are indicated for each function. The high correlation coefficient indicates that the function may be considered exponential as a first approximation.
tween pre-ictal pressure and pressure during the recovery phase (Ap) was plotted as a function of time (Fig. 4). A straight line drawn through the data points describes the data well (correlation coefficients (r 2) above 0.90). The slope of this fine may be defined as the time required to decrease the pressure by a factor of 10. This duration may be termed the pressure-time index (see also Marmarou et al. 1978, for pressure-volume index). The time constant, which is the time required for
502
A.J. GABOR ET AL.
the pressure to d r o p to 37% of its original value, was calculated. T h e time c o n s t a n t was calculated as the reciprocal of the slope of the straight line r e p r e s e n t i n g the n a t u r a l log of the pressure difference with respect to time. T h e m e a n time constant for four of the seizures was 85.5 _+ 8.73 sec. T h e time c o n s t a n t for the brief seizure was 112.4 sec.
A.
EEG
B. SP,KES IIIIIIIIIIIIIIIIIIII
I
I
C. [-
I
"I
Seizure
I
u
Sec.
Fig. 5. Diagrammatic representation of seizure-ICP relationship. A, B: a spike during the seizure is considered a unitary event in time. C: bar graph depicting number of spikes occurring at 5 sec intervals. D: a constant increment of ICP with each spike is postulated. Following each increment ICP, the ICP declines exponentially toward prespike levels. The next spike increments the ICP from the residual value (o) existing at the time the spike occurred. For convenience of calculation, n spikes in a 5 sec period was considered to be an increment of ICP n times the value of a unitary ICP increment. For each bin of the histogram in C, a scaled ICP pulse is added to the residual pressure and the sum yields a function predicting ICP as a function of time. The predicted ICP function (A p = P~F', e -(T-xn)/Tc) is plotted. Ps = a scaling function (mm Hg/spike); Fn = average number of spikes/sec for 5 sec interval; T,: = time constant of the recovery phase; ( T - T n ) = an interval of time, referenced to the beginning of the exponential weighting function (Bower and Schultheiss 1958). A, B, C, and D are drawn on the same time base.
Ictal phase In c o n t r a d i s t i n c t i o n to the recovery phase, the changes of pressure as a function of time d u r i n g the ictal phase, d i d not fit a simple e x p o n e n t i a l function. To account for the time course of the I C P d u r i n g the ictal phase, the electrical events d u r i n g the seizure were studied. The f u n d a m e n t a l unit of ictal activity available for m e a s u r e m e n t is the E E G spike. The voltage of the spikes was relatively c o n s t a n t t h r o u g h o u t the ictus. F o r p u r p o s e s of this study, the a s s u m p t i o n was m a d e that each spike represents an i n d e p e n d e n t unit of p a t h o p h y s i o l o g i c seizure activity that i n c r e m e n t s the ICP. Viewed from this perspective, the pressure change d u r i n g the ictal phase could be due to short d u r a t i o n i n c r e m e n t s of I C P occurring with each spike a n d s u m m a t i n g over time (Fig. 5). E a c h i n c r e m e n t of I C P would then decrease tow a r d pre-ictal levels with a time course d e s c r i b e d for the recovery p h a s e until the occurrence of the next spike and pressure increment. C o m p l e t e p r e d i c t e d pressure curves could be d r a w n by s u m m a t i n g i n c r e m e n t s of I C P for every spike during the seizure (Fig. 6). The c o n t i n u o u s function thus g e n e r a t e d was scaled b y visual app r o x i m a t i o n to best fit the m e a s u r e d e x p e r i m e n t a l data. This scaling d i m e n s i o n e d each i n c r e m e n t of I C P in m m H g / s p i k e . The relationship of the p r e d i c t e d pressure curves to m e a s u r e d values of I C P are shown in Fig. 6 for all 5 seizures. It should be n o t e d that the m e a s u r e d values of I C P vary from the p r e d i c t e d values b y no m o r e than 2 m m Hg. This difference is within the range of error a t t r i b u t a b l e to the technique of m e a s u r e m e n t a n d r e c o r d i n g of data. In each case, the m a t h e m a t i c a l function closely a p p r o x i m a t e s the rate of increase of intracranial pressure, the p e a k i n t r a c r a n i a l pressure, the phase difference between m a x i m a l spike frequency a n d m a x i m a l ICP, and the rate at which the I C P returns to pre-ictal values after t e r m i n a t i o n of the seizures.
Discussion
The t e m p o r a l association between seizures a n d increased i n t r a c r a n i a l pressure is well d o c u m e n t e d in the literature. Previous w o r k r e p o r t e d on the
I N T R A C R A N I A L PRESSURE D U R I N G SEIZURES
503
A
B
i4 Sp/s to 6
Sp/s i
AP ij 0J • -~
6
,6o
-:;o 6
~6o
,bo
'
Sec.
'
26o
'
a~
Sec.
D
Sp/sto
SP/Sl2 AP4
AP4
2 -2b 6
too '
~ -2o
~ 0
;
!
...............
Sec.
,6o Sec.
LB~
F"
,o]
:t
'
{IP41 J
•
2
•
o] i"
-20
'1
I
0
" !
I00
|
/
Z00
See. Fig. 6. The ICP associated with the patient's 5 seizures are plotted as a function of time. Measured ICP is indicated by symbols ©, e, r,, A II. The calculated ICP function is represented by solid lines. Scaling functions (P~, Fig. 5) have an average value of 0.07 + 0.01 m m Hg/spike,
mechanisms directly responsible for the increased pressure such as documenting causes for increased cerebral blood flow. None have attempted to quantitate the transduction of the physio-pathologic event (the epileptogenic event) to changes in
ICP. The present study describes a model for extrapolating electrical seizure activity to ICP changes. The model accounts for many of the empirically observed characteristics of the relationship between seizure activity and ICP. These
504
include the following: (1) Seizure onset before onset of increased ICP. (2) Rate of increase in ICP. (3) Temporal relationship between peak spike frequency and peak ICP. (4) Temporal relationship between peak ICP and seizure termination. (5) Time course of ICP following termination of the seizure. Perhaps the most striking aspect of this study is that the time course of the ICP changes can be predicted on the basis of only two measurable values (the electrical spike frequency of the ictal phase and the time course of the ICP recovery phase). The time course of the recovery phase ~s presumed to represent a pressure response of the intracranial physical system to small changes in intracranial volume. The return of pressure to pre-ictal values is assumed to be due to normal CSF pressure-volume dynamics, i.e., it is a passive response to a previous increase in CSF volume. The time course of the ICP during the recovery phase is primarily dependent upon outflo~ resistance and the compartmentalization of ttae increased intracranial volume. Outflow resistance may be elevated due to restrictions of CSF flow or jugular/venous sinus obstruction. Transfer of fluid to the intracerebral compartment (edema) from the vascular compartment or CSF will also alter the time course of ICP during the recovery" phase. The presence of edema may be considered a form of increased outflow resistance. It is presumed that the pressure-time relationship of the recovery phase (Fig. 4) is related to the pressure-volume relationships described in other experimental studies (Marmarou et al. 1978: Sklar and Elashvili 1977). The change of ICP undoubtedly reflects changes in intracranial xolume originating in the vascular system. The distribution of the fluid within the cranium remains uncertain. In the case described here, there was a rapid return of the pressure to pre-ictal values. This suggests that most of the increased volume was within the intravascular compartment (vasodilation) rather than being distributed in the extravascular space (edema). If a large proportion of fluid were extravascular, such as might occur with prolonged
A.J. GABOR ET AL.
seizures, the cerebral compliance could decrease for extended periods of time. This in turn would increase the potential for greater peak pressures with each subsequent seizure. The data provided here show that the pressure-time function is not the same for the ictal and recovery phase. The time course of the change in pressure during the ictal phase was quantitatively related to the ongoing epileptogenic electrical activity. It may be argued that each ictal event (represented by the EEG spike) results in an average increment of intracranial pressure. This average increment of pressure will be dependent on multiple factors and is likely to vary between patients and between seizures in the same patient. Following each increment of ICP, the ICP declines toward pre-ictal levels. Each pressure increment adds to the residual pressure existing at the time the spike occurred. The model incorporates these concepts and predicts the following relationships between electrographic events and ICP: (1) The rate of pressure change will be variable during the ictal phase. (2) The rate of pressure change will be proportional to the spike frequency over short intervals of time (0.1 time constant). (3) The peak pressure will occur and be maintained when the spike frequency is sufficiently high to counteract the decline of ICP toward preictal levels. (4) The peak pressure will occur after the peak spike frequency with a time delay determined by the spike frequency and the time course of the recovery phase. (5) If the seizure continues beyond the time of the peak spike frequency, with spikes occurring at a lower rate, the pressure will decrease despite continuation of the ictal activity. The predicted values of ICP closely approximate the experimentally derived data. Therefore, the time course of the ICP appears to be determined by the summations of the fundamental units of abnormal synchronized activity (the epileptogenic spike) and the CSF pressure-volume dynamics existing at the time of the seizure. An average increment of ICP per spike can be calculated for each seizure.
INTRACRANIAL PRESSURE DURING SEIZURES It is presumed that most seizure activity will be accompanied by increased intracranial volume. This increased volume, however, may not be associated with an increased ICP if the system is highly compliant. A poorly compliant system will be associated with changes in pressure during a seizure and the lower the compliance the greater the pressure change. Therefore, it may be anticipated that some patients will show no evidence of change in 1CP during seizures. On the other hand, compliance will be reduced during prolonged seizures and result in much higher ICP than would be expected with brief seizures. This is exemplified by the ICP of the patient presented here following a prolonged seizure prior to anticonvulsant therapy. Using the model described here, the ICP after 10 min of continuous seizure activity can be predicted. Assuming an average spike frequency of 12 spikes/sec, a recovery phase time constant of 90 sec, and a pre-ictal pressure of 10 mm Hg, the ICP after 10 min of seizure activity would be 25 mm Hg or 338 mm H20. In the patient described here, pressures ranged from 230 mm H20 to 41)5 mm H20 during a continuous seizure of unknown duration. The risk of patients developing high ICP due to prolonged seizures will be significantly increased if patients are therapeutically paralyzed and the occurrence of seizure activity is difficult or impossible to recognize by clinical observation.
Summa~ A comatose 31-year-old male with presumed viral encephalitis and frequent partial motor seizures was paralyzed with pancuronium in an attempt to reduce recurrent elevation of intracranial pressure (ICP) associated with each seizure. ICP was continuously monitored with a Richmond Bolt and 5 electrographic seizures originating in the left frontal area were recorded. Each ictal episode was associated with stable blood pressure and an increase of ICP. The average seizure duration was 78 + 17 sec (mean :t S.D.) and the average maximum increase of ICP above baseline during the seizures was 6.5 + 0.6 mm Hg with average peak ICP of 16.0 + 0.86 mm Hg. A
505 simple mathematical model predicts the rate of increase of ICP, the peak ICP, the phase difference between maximum spike frequency and maximum ICP, and the rate at which ICP returns to pre-ictal values after termination of the seizure. The predicted values of ICP closely approximate the experimentally derived data. Therefore, the time course of the ICP appears to be determined by the frequency of the fundamental units of abnormal synchronized activity (the epileptigenic spike) and the CSF pressure-volume dynamics existing at the time of the seizure. An average increment of ICP per spike can be calculated for each seizure. The model also predicts that patients may develop high ICPs due to prolonged seizures. Prolonged unrecognized seizures may occur in patients who are therapeutically paralyzed as demonstrated by the case described here.
R6sum6 Pression intracrgmienne au cours de crises bpileptiques
Un sujet male de 31 ans dans un 6tat comateux pr6sentant une enc6phalite virale pr6sum6e et de fr6quentes crises 6pileptiques motrices partielles a 6t6 paralys6 au pancuronium afin de tenter de diminuer l'616vation r6p6t4e de la pression intracr~nienne (ICP) associ6e h chaque crise. L'ICP 6tait sous surveillance continue h l'aide d'un Richmond Bolt et 5 crises 61ectrographiques ayant leur origine dans l'aire frontale gauche ont 6t6 enregistr6es. Chaque 6pisode ictal 6tait accompagn6 d'une pression art6rielle stable et d'une augmentation de I'ICP. La dur6e moyenne de la crise a 6t6 de 7 8 + 17 sec (moyenne+ E.T.) et la moyenne de l'augmentation maximale de I'ICP au-dessus de la ligne de base au cours des crises, de 6,5 + 0,6 mm Hg avec un pic moyen de I'ICP h 16,0 + 0,86 mm Hg. Un mod61e math6matique simple permet de pr6dire le taux d'augmentation de I'ICP, le pic ICP, la diff6rence de phase entre la fr6quence maximum de pointes et le maximum de I'ICP, et la vitesse h laquelle I'ICP revient h des valeurs pr6ictales apr6s la fin de la crise.
506
Les valeurs pr~dites pour I'ICP sont une approximation tr6s proche des valeurs obtenues exp6rimentalement. Le d6cours temporel de I'ICP apparait donc d6termin~ par la sommation des unit6s fondamentales de l'activit6 anormale synchronis6e (pointe 6pileptog/me) et la dynamique pression-volume du liquide c6phalorachidien existant au moment de la crise. Une valeur movenne de l'augmentation de I'ICP par pointe peut ~tre calcul6e pour chaque crise. Le mod61e pr6dit aussi que les patients peuvent d6velopper des ICP 61ev6es lors de crises prolong6es. Des crises prolong6es non reconnues peuvent survenir chez des patients qui sont paralys6s th6rapeutiquement comme le d6montre te cas d6crit ici. References Bower, J.L. and Schultheiss, P.M. Weighting functions. In: Introduction to the Design of Servomechanisms Wiley, New York, 1958: 48-50. Brodersen, P., Paulson, O.B., Bolwig, T.G., Rogon, Z.E., Rafaelsen, O.J. and Lassen, N.A. Cerebral hyperemia in electrically induced epileptic seizures. Arch. Neurol. (Chic.), 1973, 28: 334-338.
A.J. GABOR ET AL. lngvar, D.H. CBF and electrical activity. Stroke, 1973, 4: 359. Marienne, J.P., Robert, G. and Bagnat, E. Post-traumatic acute rise of ICP related to sub-clinical epileptic seizures. Acta neurochir. (Wien), 1979, 28 (Suppl.): 89-92. Marmarou, A., Shulman, K. and Rosende, R.M. A nonlinear analysis of the cerebrospinal fluid system and intracranial pressure dynamics. J. Neurosurg., 1978, 48: 332-344. Miller, J.D. Volume and pressure in cerebrospinal axis. Clin. Neurosurg., 1975, 22: 76-105. Minns, R.A. and Brown, J.K. Intracranial pressure changes associated with childhood seizures. Develop. Med. Child. Neurol., 1978, 20: 561-569. Plum, F., Posner, J.B. and Troy, B. Cerebral metabolic and circulatory responses to induced convulsions in animals. Arch. Neurol. (Chic.), 1968, 18: 1-13. Sakai, F , Meyer, S., Naritomi, H. and Hsu, M. Regional cerebral blood flow and EEG in patients with epilepsy. Arch. Neurol. (Chic.), 1978, 35: 648-657. Sklar, F.H. and ElashvilL I. The pressure-volume function of brain elasticity. J. Neurosurg., 1977, 47: 670-679. Tans, J.TJ. and Poortviet, D.CJ. Intracranial volume-pressure relationship in man. Part l: Calculation of pressure-volume index. J. Neurosurg., 1982, 56: 524-528. Tsementzis, S.A., Gillingham, FJ. and Hitchcock, E.R. The effect of focal twitching on the intracranial pressure during paralysis and mechanical ventilation. Ann. clin. Res., 1979, 11: 253-257. White, P.T., Grant, P., Mosier, J. and Craig, A. Changes in cerebral dynamics associated with seizures. Neurology (Minneap.), 1961, 11: 354-361.