Cerebral Blood Flow in Children with Intractable Epilepsy Hiroshi Shimizu, MD, Tetsuzo Tagawa, MD, Yasuyuki Futagi, MD and Junko Tanaka, MD A good correlation of Doppler internal carotid blood velocity determinations with hemispheric blood flow measurements by the 133-xenon inhalation method was demonstrated in 14 epileptic patients without abnormal CT findings. The highest correlation was seen between the flow of gray matter (F1) and the internal carotid end-diastolic velocity (d) (left side r = 0.841, right side r = 0.817). As end-diastolic velocity (d) well correlated with the value obtained by the 133-xenon inhalation method, the d value was compared between 77 healthy children and 13 patients with intractable epilepsy. The mean d value of both internal carotid arteries in patients was 16.7 ± 2.9 mm (mean ± SD), and that of healthy children 20.9 ± 4.3 mm, the difference being statistically significant. The low cerebral blood flow in patients might be due to multiple antiepileptic drugs administered and/or mental retardation and cerebral hypofunction related to seizures. Shimizu H, Tagawa T, Futagi Y, Tanaka J. Cerebral blood flow in children with intractable epilepsy. Brain Dev 1983; 5:36-40
Non-invasive measurement of cerebral blood flow (CBF) is very useful for clinical assessment of cerebral function at rest, especially in children. In 1960, Sat omura et al [1] reported on the usefulness of Doppler-shifted ultrasound signals to assess the carotid arterial blood flow, which was confirmed by subsequent investigators [2, 3]. On the other hand, 133-xenon inhalation method was introduced into the same armamentarium by Mallett et al [4] in 1965. The latter, however, has not been recognized yet as a reliable objective method to
From the Department of Pediatrics, Osaka Medical Center for Women & Children, Osaka (HS, YF, In; Department of Pediatrics, Osaka University Medical School, Osaka (IT). Received for publication: September 24,1982. Accepted for publication: January 12, 1983.
Key words: Cerebral blood flow, intractable epilepsy, children. Correspondence address: Dr. Hiroshi Shimizu, Department of Pediatrics, Osaka Medical Center for Women & Children, 840 Murodocho, Izumi, Osaka 590-02, Japan.
evaluate CBF since the correlation is still not adequately validated against a standard method of CBF measurement. In spite of such drawback, the 133-xenon inhalation technique allows a demonstration of changes of total CBF which agree well with CBF changes evaluated with the arteriovenous oxygen content difference method. In the present study, the authors undertook a comparative study of two methods; that is, the Doppler ultrasonic method to measure internal carotid velocity and the 133-xenon inhalation method to measure hemispheric CBF. A good correlation was demonstrated between the two methods in 14 epileptic children. Then, the Doppler internal carotid blood velocity study was applied to patients with intractable epilepsy and the results was compared with that of healthy children. Principle and Methods
A continuous wave Doppler transducer generates an ultrasound beam of a frequency of 5MHz. The ultrasound beam intersects at the center of the blood vessel at an angle of 60
5,
Table 1 Age distribution of subjects for Doppler ultrasonic assessment. Age (yrs)
8 9
Numbers of cases Patients
2 2
10
-------------L-------------
Schema of sound spectrogram. S,: initial systolic peak, S,: second systolic peak, I: incisura, D: diastolic peak, d: end-diastolic peak. Fig 1 Schema of sound spectrogram.
degrees. The sound waves are reflected from the moving blood elements. The frequency of these reflected waves is slightly different from the transmitted signal. This difference is given by the Doppler equation: 2Vf fd = - - cosO c where fd is the Doppler difference frequency; f is the transmitted frequency; V is the velocity of blood cells; C is the velocity of sound in blood (1,500 m/.sec); and 0 is the angle between the sound beam and velocity vector. Our ultrasonic Doppler method involves recording the detected sound on a tape and analyzing it by sound spectrography; this method provides more reliable records than other methods. A sound spectrogram of the velocity pulse wave obtained from an artery is shown schematically in Fig I, where A (mm 2 ) is the area surrounded by the maximum velocity curve in one cardiac cycle and L (mm) is the duration of one cardiac cycle. The A/L (mm) value is equivalent to the mean value of the maximum flow velocity and the value of d is the end-diastolic blood velocity. Patients were studied in the supine position. The probe of a Doppler ultrasonic velocity detector was positioned over the internal carotid artery in such a way that beams from the probe intersecte the blood stream at an angle of 60 degrees. Then, the blood velocity, A/L, was 30 cm/sec and the d value 12 mm. The hemispheric CBF was measured by the 133-xenon inhalation technique by Obrist et al [5]. Discussion of the 133-xenon inhalation method is found in other papers [6, 7].
11
12
3
13
0 2
14 15 16 17 18 Total
Normal controls
6 12 12 11 12
0 0
12 5 5 1 0
13
77
The patients inhaled 133-xenon mixed with air for one minute by means of a face mask and a rebreathing system. Five mCi of 133-xenon was used in each of the studies. The one minute inhalation period was followed by 10 minutes during which the patient breathed normal air. With the children lying supine, the gamma radiation was recorded by 18 scintillation detectors placed in parallel at right angles to the lateral surfaces of the patient's head. The results presented here are based on the Initial Slope Index(ISI). similar to the one proposed by Risberg et al [7], and flow of gray matter (Fl) contrived by Obrist et al [5] . Subjects First, in order to compare the internal carotid velocity measured by the Doppler ultrasonic method and mean hemispheric CBF by the 133xenon inhalation method, 14 epileptic patients were studied. They were 10 males and 4 females, aged between 5 years and 15 years. In all cases, seizures were well controlled, and nobody had abnormal CT findings nor neurological abnormalities. Second, for Doppler ultrasonic assessment, 77 healthy children and 13 patients with intractable epilepsy were examined. Healthy children were 38 males and 39 females, aged between 8 years and 18 years, averaging 11.9 years (Table 1). The patients with intractable Shimizu et al. CBF in epileptic children 37
00 .....
\0
..... .....
~
.v,
~
::!
....'"
;:!
'l;:j
'"-.:: '" C
t:::l
Roo
S·
~
b:l
00
.....
SGE SGE
15:6
8:9
14:7
8:3
10:0
0:4
0:6
0:6
0:11
2:2
1:8
F
M
M
M
M
F
9
10
11
12
13
SGE
Anticonvulsant drugs
MR (lQ 55) PB, CBZ, VPA, CZP, ESM, AZA MR (lQ 59) PRM, AZA, DZP, PB 13 VPA 35 VPA
Tonic,2/D
Tonic, lID, absence, lID
Normal
Normal
MR (IQ 56) CBZ, AZA, VPA, VP A 72 CBZ 4.5 Slight atrophy ESM, CZP ESM 59.5
Normal
MR (lQ 60) PB, VPA, NZP
Astatic, 1-2/D Tonic,1-2/D, myoclonic, 6-7/D
MR (lQ 39) PRM, AZA, CBZ, PB 12.3 CBZ 3.5 Slight NZP, VPA PRM 5.6 atrophy
Microcephalus PRM, VPA, CBZ, PB 26 CBZ 4 (IQ 39) DZP,AZA VPA 77
Tonic, 4-5/D, absence, 10-20/D
Focal motor, 10-20/D
Slight atrophy
Normal
PB 88 PHT 20 VPA86
- (lQ 92) PB, PHT, VPA, AZA
Focal motor, 1/W, grand mal, 1/2W
Normal
MR (lQ 52) PRM, PHT, VPA, PB 29 PHT 3 clobazam VPA 114
Normal
Myoclonic, 1-2/D
PB 6.8 PHT 7
Normal
MR (lQ 39) PB, PHT, CBZ, NZP, AZA
MR (lQ 55) PRM, PHT, CBZ, PB 38 CBZ 1 NZP, AZA PHT 16
Slight atrophy
Normal
Normal
CT
Tonic,l/W, absence, 2/D
Tonic,l/W
PB 28 CBZ 3 PHT 10
MR (lQ 45)" PB, PHT, CBZ, ESM,CZP
Myoclonic, 20-3 OlD, absence,1-2/D
j
PB 43 PHT 5 CBZ 1.5
Serum concentration (pglml)
MR (lQ 35) PRM, PHT, VPA, PB 13 PHT ESM, CZP, AZA ~PA66
MR (lQ 30) PB, PHT, CBZ, clobazam
Underlying diseases
Astatic, 3-4/W
Tonic,1-2/W
Seizure types & frequency
27.5
20.1
Slow sp-w
Slow sp-w
26.4 29.0
Slow sp-w Irregular sp-w
28.7
24.5
Focal spike, asymmetry
Slow sp-w
24.0
26.7
24.7
32.8
27.8
30.7
29.1
AIL (mm)
17.5
18.0
17.5
13.5
16.0
13.5
12.3
17.3
12.8
20.0
20.5
20.5
18.0
d (mm)
Internal carotid velocity
Focal spike
Slow sp-w
Slow sp-w
Irregular sp-w
Irregular sp-w
Slow sp-w
Slow sp-w
EEG
AZA: acetazolamide, CBZ: carbamazepine, CZP: clonazepam, DZP: diazepam; ESM: ethosuximide, NZP: nitrazepam, PB: phenobarbital, PHT: phenytoin, PRM: primidone, VPA: sodium valproate, M: male, F: female, SGE: secondary generalized epilepsy, MR: mental retardation, sp-w: spike and wave.
11:2
SGE
SGE
SGE
Partial epilepsy
8
12:3
11:0
M
SGE
SGE
SGE
SGE
SGE
SGE
7
14:8
18:2
12:2
12:2
9:11
9:8
Age at examination Epilepsy (year: types month)
1:6
7:0
0:6
0:5
0:4
0:2
Age at onset (year: month)
M
M
M
M
M
M
Sex
Clinical descriptions of 13 epileptic children.
6
5
4
3
2
1
Case
Table 2
epilepsy were 11 males and 2 females, aged between 8 years and 18 years, averaging 12.3 years. Seizure types in these patients were variable and epilepsy types were secondary generalized epilepsy in 12 cases and partial epilepsy in one. All cases had no neurological abnormalities except mental retardation. They had abnormal EEG findings and at least 3 antiepileptic drugs have been prescribed (Table 2). Results
The correlations between internal carotid velocity measured by the Doppler technique and mean hemispheric CBF measured by the 133-xenon inhalation method in 14 wellcontrolled epileptics are shown in Table 3. The highest correlation was seen between the Fl value and the d value (left side r =0.841, right side r =0.817) (Table 3). The other parameters showed a lower correlation. As end-diastolic velocity (d) well correlated with the value obtained by the 133-xenon inhalation method, the d value was compared
Table 3 The correlations between AIL and d values in the internal carotid artery measured by the Doppler technique and in hemispheric CBF measured by the 133-Xe inhalation method (n = 14).
Xe·inha· lation Doppler d AIL
lSI
Fl It
rt
0.8413 0.8170 (p < 0.001) 0.7748 0.7110 (p < 0.01)
It
rt
0.7412 0.7284 (p < 0.01) 0.7095 0.6110 (p < 0.02)
Table 4 AIL and d values of healthy children and patients with intractable epilepsy.
No of subjects Age (mean) AIL (mm) d(mm)
*p
< 0.01.
Normal subjects
Intractable epilepsy
77 8-18 y (11 yrs 10 mos) 32.6 ± 6.0 20.9 ± 4.3
13 8-18 y (12 yrs 3 mos) 27.1 ± 3.3* 16.7 ± 2.9*
between healthy children and patients with intractable epilepsy. The mean d value of both internal carotid arteries in healthy children was 20.9 ± 4.3 mm, and that of patients with intractable epilepsy 16.7 ± 2.9 mm, the difference being statistically significant (Table 4). Discussion
In this study, good correlation coefficients were found between mean hemispheric CBF and some parameters of internal carotid Doppler velocity in spite of the fact that arterial diameters were not taken into account. Risberg et al [8] also reported the high correlation between CBF measured by the 133-xenon inhalation method and internal carotid enddiastolic velocity by the Doppler ultrasonic method in adult patients, and the correlation coefficients were 0.83 (left side) and 0.88 (right side). In the present study, correlation coefficients between CBF measured by the 133xenon inhalation and end-diastolic velocity (d) were 0.841 (left side) and 0.817 (right side). The mean value of the maximum flow velocity values (AIL) showed a less close relationship to hemispheric CBF (r =0.775 (1t), r =0.711 (rt)). Based on these data, the end-diastolic velocity and the mean value of the maximum flow velocity of the internal carotid artery (AIL) can allow estimation of the hemispheric blood flow. According to Ingvar [9], CBF in adult patients with focal cortical epilepsy was by and large normal interictally, and over focal regions, where intermittent paroxysmal discharges are often generating on EEG, a reduced flow was recorded in the interictal stage. In the ictal stage, very high flows were recorded over the discharging focal region [9]. Other authors [10] measured CBF in the interictal state in the patients with intractable epilepsy and demonstrated that it was statistically lower than that of healthy children and that of children with well controlled epilepsy. The subjects in our study had intractable variable seizures and suffered from mental retardation, and their EEGs showed many paroxysmal epileptic discharges from multiple foci with a slow basic rhythm. In addition, they were prescribed many antiepileptic drugs. Some authors [11, 12] reported that the frequency of basic EEG waves varied in parallel with CBF and a reduction of CBF resulted in a decrease of the EEG Shimizu et al. CBF in epileptic children 39
frequency. It was also reported that CBF of patients with mental retardation was lower than that of healthy children [13, 14]. Antiepileptic drugs (phenobarbital [1S] and diazepam [16]) significantly reduced CBF and cerebral oxygen uptake. The low CBF found in our patients with intractable epilepsy might be due to either antiepileptic drugs administered or mental retardation or cerebral dysfunction related to seizures or all in combination.
References 1. Satomura S, Kaneko Z. mtrasonic blood rheo· graph. Proc of the 3rd International Conference of Med Electronics. London. 1960:254-7. 2. Shimizu H, Sumi K, Sugita T, et al. Cerebral blood flow changes in orthostatic dysregulation fainters. Brain Dev (Tokyo) 1982;4:21-6. 3. Futagi Y, Shimizu H, Mimaki T, et al. Internal carotid blood flow velocity in children with cerebral palsy by Doppler ultrasound method. Brain Dev (Tokyo) 1982;4:27-33. 4. Mallett BL, Veall N. The measurement of regional cerebral clearance rates in man using xenon133 inhalation and extracranial recording. Clin Sci 1965;29:179-91. 5. Obrist WO, Thompson HKJr, Wang HS, et at. Regional cerebral blood flow estimated by 133xenon inhalation. Stroke 1975;6:245-56. 6. Obrist WO, Thompson HKJr, King CH, .et at. Determination of regional blood flow by ·inhalation of 133-xenon. Circ Res 1967;20:124-35. 7. Risberg J, Ali Z, Wilson EM, et al. Regional
40 Brain & Development, Vol 5, No 1, 1983
8. 9.
10.
11.
12.
13.
14. 15.
16.
cerebral blood flow by 133-xenon inhalation. Preliminary evaluation of an initial slope index in patients with unstable flow compartments. Stroke 1975;6:142-8. Risberg J, Smith P. Prediction of hemispheric blood flow from carotid velocity measurements. Stroke 1980;11:399-402. Ingvar DH. rCBF in focal cortical epilepsy. In: Langfitt TW, McHenry LCJr, Reivich M, et a1, eds. Cerebral circulation & metabolism. New York: Springer-Verlag, 1975:361-4. Kennedy C, Anderson W, Sokoloff L. Cerebral blood flow in epileptic children during the interseizure period. Neurology (supplement) 1958;1: 100-5. Darrow CW, Graff CG. Relation of electroencephalograms to photometrically observed vasomotor changes in the brain. J Neurophysiol 1945 ;8:449-61. Parkes JD, James 1M. Electroencephalographic and cerebral blood flow changes following spontaneous subarachnoid haemorrhage. Brain 1971; 94:69-76. Futagi Y. Internal carotid velocity in children with developmental disabilities by Doppler ultrasound method (in Japanese). No To Hattatsu (Tokyo) 1983;.in press. Garfunkel JM, Baird HW, Ziegler J. The relationship of oxygen consumption to cerebral functional activity. J Pediatr 1954;44:64-72. Hagerdal M, Keyhah M, Perez E, et al. Additive effects of hypothermia and phenobarbital upon cerebral oxygen consumption in the rat. Acta Anaesthesiol Scand 1979;23:89-92. Maekawa T, Sakabe T, Takashita H. Diazepam blocks cerebral metabolic and circulatory responses to local anesthetic-induced seizures. Anesthesiology 1974;41:389-91.