Original Articles
Changes in CSF Neurotransmitters During the First Year of Life Hitoshi Yamamoto, MD
Changes in cerebrospinal fluid neurotransmitter metabolites in 25 children younger than 1 year of age were analyzed to assess maturation of the central nervous system and were compared to cerebrospinal fluid from older children and adults. Significant inverse correlations (P < .05) with aging were observed for tryptophan, 5-hydroxytryptophan, 5-hydroxyindoleacetic acid, kynurenine, tyrosine, dopa, dopamine, dihydroxyphenylacetic acid, homovanUlic acid, norepinephrine, and 3-methoxy-4-hydroxyphenyiglycol concentrations. There were no significant differences observed with respect to age in the cerebrospinal fluid serotonin, 3-hydroxykynurenine, and 3-methoxytyramine concentrations. This study suggests that changes in the major cerebrospinal fluid neurotransmitters occur with increasing age during the neonatal period. Because these findings are preliminary, additional patients require study. Yamamoto H. Changes in C S F neurotransmitters during the first year of life. Pediatr Neurol 1991;7:406-10.
Introduction The first year of life is a period of very rapid development and is of special interest in measuring cerebrospinal fluid (CSF) levels of neurotransmitter metabolites. CSF levels of neurotransmitter metabolites have been widely studied to ascertain neurotransmitter turnover in the central nervous system (CNS). Problems remain regarding the degree to which the peripheral levels are correlated with brain levels and whether alterations occur in particular neuroiogic disorders [1-3]. The fact that medications effective in ameliorating symptoms in some neurologic disorders have specific effects on CSF neurotransmitters, as well as altered CSF neurotransmitter metabolite levels in different diagnostic groups, has encouraged continued investigation of neurotransmitters in neurologic disorders of childhood. C S F concentrations of h o m o v a n i l l i c acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA) have
From the Department of Pediatrics; St. Marianna University School of Medicine; Kawasaki 213, Japan.
406 PEDIATRIC NEUROLOGY Vol. 7 No. 6
been determined in adult controls and in adults with neurologic disorders [2,4,5]. In adult neurologic disorders, an age effect has been reported for CSF HVA and possibly for CSF 5-HIAA [6,7]. The availability of the precursor amino acids in brain has been believed to affect central serotonin synthesis [8,9] and, to a lesser extent, dopamine and norepinephrine synthesis 110]. The issue of precursor control of neurotransmitter synthesis in children can be studied by examining correlations between precursors and metabolites. The ontogenetic pattern of precursor levels is also of interest, regardless of the extent to which these levels influence the rate of neurotransmitter synthesis. The usefulness o f C S F neurotransmitter metabolite measures in infants has been limited by a lack of normal control data. In disorders believed not to disturb CNS neurotransmitter metabolism, CSF obtained by diagnostic lumbar puncture has been analyzed for neurotransmitters [3,11]. Although an age effect for HVA and possibly 5H I A A has been observed in some of these studies [6,12], mean values for discrete age intervals have been presented in only a few reports [ 13,141. Individual variability of C S F neurotransmitter metabolites, including kynurenine (KYN) and 3-hydroxykynurenine (3-OHKYN), in neonates has not been studied. In the present study, the levels of tryptophan (TRP), 5-hydroxytryptophan (5-HTP), serotonin (5-HT), 5-HIAA, KYN, 3-OHKYN, tyrosine (TYR), L-dihydroxyphenylalanine (L-DOPA), d o p a m i n e (DA), dihydroxyphenylacetic acid (DOPAC), HVA, norepinephrine (NE), 3-methoxy-4-hydroxyphenyl glycol (MHPG), and 3-methoxytyramine (3MT) were measured in the CSF of neurologically normal infants, children, and adults in order to establish normal values as well as to characterize interrelationships among these compounds. The tryptophan and tyrosine metabolic pathways are illustrated in Figure 1.
Methods CSF was obtained from 25 neurologically normal infants (newborn to 11 months), 7 children (3-7 years), and 9 adults. Samples were obtained from infants and children who required lumbar puncture for diagnosis of possible meningitis and from adults with headache or
Communications should be addressed to: Dr. Yamamoto; Neurology Service (V-151); VA Medical Center; 3350 La Jolla Village Drive; San Diego, CA 92161. Received March 7, 1991; accepted July 1, 1991.
/, TRP
/-
KYN
Each of the compounds was analyzed for the 5 age groups using a standard computerized analysis of variance (ANOVA) statistical procedure (Statistical Analysis System).
) 3-OHKY
)5-HTP
) 5 - H T --~-~ 5 - H I A A
Results
3-MT TYR
) L-DOPA
DOPAC
)DA '~NE
)HVA
--~-~ MHPG
Figure 1. The tryptophan and tyrosine metabolic pathways. Abbreviations: TRP, tryptophan; 5-HTP, 5-hydroxytryptophan; 5-HT, serotonin; 5-HIAA, 5-hydroxyindoleacetic acid; KYN, kynurenine; 3-OHKYN, 3hydroxykynurenine; TYR, tyrosine; L-DOPA, L-dihydro.ryphenylalanine; DA, dopamine; DOPAC, dihydroxyphenylacetic acid; HVA, homovanillic acid; NE, norepinephrine; MHPG, 3-methoxy-4-hydroxyphenyl glycol; 3-MT, methoxytyramine. neuropathy; the adults and children were later shown to be neurologically normal. The infants' medical histories were examined carefully and none had any of the following signs: prematurity, seizures, asphyxia, hydrocephalus, sepsis, intraventricular hemorrhage, or CNS medications. CSF was collected by lumbar puncture while the patient was in the lateral decubitus position and was immediately frozen at -70°C for later analysis. The total volume of CSF removed from each infant or child was typically 3 ml; 3-5 ml samples were taken from adult patients. CSF was collected in 2 aliquots (from infants and children) or 3 aliquots (from adults). The last aliquot (0.5 ml) was always used for analysis. Analysis of neurotransmitter metabolites was accomplished by direct injection of CSF (50 lad onto a computer-controlled, high-performance, liquid chromatographic (HPLC) analyzer. This instrument, the Neurochem neurochemical analyzer (ESA, Bedford, MA), consists of a gradient HPLC system and 16 high-sensitivity coulometric electrochemical detectors. The concept and inherent advantages of this multielectrode HPLC system have been described elsewhere [15,16]. The gradient profile, column, and detector potentials used in this study were similar to those previously described [16]. The mobile phase solutions were obtained directly from ESA (Bedford, MA) and had the following compositions: Mobile phase A - 0 . 1 sodium phosphate with 10 mg/L sodium dodecyl sulfate at pH 3.35; Mobile phase B - 5 0 % methanol/water with 50 mg/L sodium dodecyl sulfate at pH 3.45. All standards were obtained from Sigma Chemical Company (St. Louis, MO). Table 1.
The patients were placed into 5 age groups for analysis: Group 1 : 5 newborns younger than 1 month of age; Group 2 : 1 0 infants 1-5 months (mean age: 3.7 mos); Group 3 : 1 0 infants 6-11 months (mean age: 8.1 mos); Group 4 : 7 children 3-7 years; and, Group 5 : 9 adults. The variations in the individual CSF neurotransmitter metabolite concentrations with age are listed in Tables 1 and 2. The levels of CSF TRP, TYR, L-DOPA, and 5-HIAA were highest in group 1 and decreased significantly with increasing age (P < .05). No significant difference was observed between the levels in groups 4 and 5. The levels of CSF 5-HTP were highest in group 1 and decreased significantly with increasing age (P < .05); however, there were no significant differences between the values in groups 2 and 3 or in groups 4 and 5. The levels of CSF KYN and DA in infants younger than 1 year of age were higher than those of the older children and after 1 year of age approached the adult levels. The concentrations of CSF MHPG, DOPAC, and NE decreased significantly in group 3 (P < .05) and were reduced to near adult levels after 6 months of age. CSF HVA concentrations also decreased significantly (P < .05) during the first year of life and were still above adult levels in group 4. The levels of CSF 5-HT, 3-OHKYN, and 3MT did not differ significantly with increasing age. Discussion
Concentrations of neurotransmitter metabolites in CSF are determined by the rates of synthesis, release, and degradation of the parent compounds as well as the rate and efficiency of elimination of metabolites from brain
Changes in CSF TRP metabolites with age* Group 1 Newborns (N = 5)
Group 2 1-5 mos (N = 10)
Group 3 6-11 mos (N : 10)
882.9 _+ 101.6
739.2 + 68.0 +
610.4 _+ 105.6 ~
337.6 _+55.2 §
289.1 _+35.2
5-Hydroxytryptophan
2.7 + 0.2
1.9 + 0.4 +
1.9 _+0.5
1.2 + 0.3 §
1.1 _+0.3
Serotonin
.04 + .02
.04 + .01
.05 _+.02
.04 + .02
.05 + .02
5-Hydroxyindoleacetic acid
130.4 + 6.9
81.0 + 11.1'
48.2 _+6.6 :~
20.3 _+5.4 §
18.3 + 7.3
Kynurenine
41.8 _+7.0
38.6 -+ 9.9
37.9 + 8.3
12.5 + 4.6§
11.7 + 7.2
3-Hydroxykynurenine
0.15 + .04
0.12 -+ .03
0.14 _+.04
0.13 + .03
0.14 _+.07
Compound Tryptophan
Group 4 3-7 yrs (N = 7)
Group 5 Adults (N = 9)
* All data expressed as ng/ml CSF (mean + S.D.). + Significantly different from Group 1, P < .05. Significantly different from Group 2, P < .05. § Significantly different from Group 3, P < .05.
Yamamoto: CSF Neurotransmitters in Neonates
407
Table 2.
Changes in CSF tyrosine metabolites with age*
Compound
Tyrosine
Group 1 Newborn (N = 5)
Group 2 1-5 mos (N : 10)
Group 3 6-11 mos (N = 10)
Group 4 3-7 yrs (N : 7)
Group 5 Adults (N = 9)
5,525.3 _+1,114
4,787 +_1,589
4.190.3+ 1,254.3
2.201 _+770§
1.877 + 203
L-Dihydroxyphenylalanine
0.76 -+0.32
0.54 i 0.14+
0.35 _+0.12:~
0.18 _+.05§
0.2 _+0.1
Dopamine
0.43 + 0.11
0.49 _+0.16
(I.4 + .07
0.15 + .04~
0.18 + .05
Dihydroxyphenylaceticacid
0.71 _+0.15
0.51 _+0.21
0.36 -+0.14'
(t.35 +_0. t2
0.29 _+0.13
Homovanillicacid
156.9 + 16.7
100.2 + 18.7+
65.7 + 13:~
63.9 _+17.1
33.6 + 11.21[
Norepinephrine
0.48 + 0.12
(I.47 + 0.2
0.31 _+0.11;
0.3 _+.07
0.31 _+0.14
3-Methoxy-4-hydroxyphenyl glycol
22.8 + 3.1
21.9 + 3.2
13.5 + 2.6+
10.7 + 2.9
7.9 -+ 1.9
3-Methoxytyramine
0.21 _+.(12
0.2 _+.02
0.2 _+.03
(I.19 _+.01
0.18 _+.02
* All data expressed as ng/ml CSF (mean i S.D.). + Significantlydifferentfrom Group 1, P < .05. Significantlydifferentfrom Group 2, P < .05. § Significantlydifferentfrom Group 3, P < .05. ¶ Significantlydifferentfrom Group 4, P < .05. and CSE Maturational changes in storage pools, rates of tumover, transport systems, intra- and extra-neuronal metabolism, rates of CSF production, and brain and spinal column morphology all may alter CSF metabolite concentrations. In previous studies, several investigators reported that the CSF levels of TRE 5-HIAA, TYR, HVA, and MHPG in neonates and the values for neonates were considerably higher than those of older children and adults. Approximately 50% of patients in each age group had fever, but fever has been demonstrated not to affect 5-HIAA and HVA levels [17]. The compound concentrations reported here in the newborn to 1l-month-old infants are in agreement with previous reports of neurologically normal infants (Table 3); however, there is little normal control data for CSF neurotransmitter metabolites and CSF levels of KYN and 3-OHKYN have not been previously reported in neonates. Seasonal variations of CSF neurotransmitter metabolite concentration have been reported in several adult psychiatric patients and normal individuals [20,21]; however, Swedo et al. reported that there were no seasonal differences in CSF concentrations of MHPG, HVA, or 5-HIAA in a pediatric population [22]. In the present study, the time of year of the lumbar puncture varied and the possibility of seasonal variations of CSF neurotransmitter metabolite levels in neonates cannot be determined. The concentration gradient of CSF also may affect the results. A lumbar-cistemal concentration gradient exists for HVA and 5-HIAA in adults [23-25]; therefore, the concentration may depend on the volume of the sample obtained. In children, a concentration gradient of CSF monoamine metabolites has also been reported [26]. It is evident that for the purpose of comparison, samples from children and adults must be collected in precisely the same manner and metabolites must be measured in the same fraction of the collected fluid.
408 PEDIATRICNEUROLOGY Vol.7 No. 6
The levels of CSF TRP and its metabolites 5-HTP and 5-HIAA were highest in newborns (younger than 1 month) and decreased gradually with age. It is probable that the age-related changes in CSF neurotransmitter metabolites are related to differences in neurotransmitter turnover rate and the rate and efficiency of elimination from CSE The mode of metabolic clearance is changed from the newbom stage through adulthood. It is reported that the main mode of clearance is via the CSF to the choroid plexus in newboms, whereas in adults these compounds exit directly from the brain to the blood, probably across the bloodbrain barrier [27]. In the present study, the levels of CSF 5-HT did not differ with increasing age and were lower than the levels reported in previous studies [28-30]. This difference may be due to the different method of detection in this study or by the volume of CSF obtained [25,26]. The presence of 5-HT in CSF may be related t o diffusion of unmetabolized 5-HT which has been released from nerve terminals. Levels of 5-HT also depend on the degree of serotoninergic activity in the CNS. The CSF levels of KYN and 3-OHKYN in neonates have not been reported. These are metabolites of TRP that are excitotoxic and are implicated in the pathogenesis of certain human neurodegenerative and seizure disorders [31,32]. In the present study, the levels of CSF KYN in neonates were higher than those of children and adults. In contrast, CSF levels of 3-OHKYN did not change with increasing age. According to Gal and Sherman, in rats the metabolism of TRP via the KYN pathway is about 45% of the metabolism of TRP to 5-HT [33]; this percentage may change with age. Except for 3MT, a significant inverse correlation was observed between age and the CSF concentration of TYR metabolites. The c o n c e n t r a t i o n of M H P G , NE, and DOPAC decreased rapidly and reached adult levels by 6-11 months of age. The L-DOPA level decreased earlier
Table 3. Literature summary of reports of CSF neurochemical levels in infants
Authors
Patients
Assay
Age (mos)
N
TRP (ng/ml)
5-HIAA (ng/ml)
TYR (ng/ml)
HVA (ng/ml)
MHPG (ng/ml) 23 ± 3 22 ± 3 16 ± 3
Present patients
Normal infants
HPLCEC
0-1 1-5 6-11
5 10 10
833 ± 102" 739 ± 68 610 ± 106
130 ± 7 81 ± 11 48+7
5,525 ± 1,114 4,788 ± 1,590 4,190 ± 1,254
157 ± 17 100 ± 19 66 ± 13
Anderson et al. [3]
Normal infants
HPLCEC/F
0-1
17
1,250 ± 100
143 ± 9
7,940 ± 1,480
184 ± 15
Langlais et al. [14]
Normal infants
HPLCEC
0-1 4-7 8-12
12 8 5
143 ±40 62 ± 16 58 ± 24
197 ± 62 148 ± 48 157 ±44
Silverstein et al. [18]
Suspicion of infection
HPLCEC
NB 0-6
19 13
104±7 85±7
129 ± 11 159 ± 9
Seifert et al. [13]
Neurologic disorder
GC-MS
0-12
14
-43**
-120"*
Rogers and Dubowitz [ 19]
Miscellaneous disorders
Fluorometry
0-12
13
99± 11
29± 14 15 ± 3 13 ± 2
* Values are mean + S.D. ** Data were not reported in numerical form (approximated). Abbreviations: 5-HIAA = 5-Hydroxyindoleacetic acid EC/F = Electrochemical detection/fluorometry GC-MS = Gas chromatography-mass spectrometry HVA = Homovanillic acid
MHPG NB TRP TYR
= = = =
in n e o n a t a l life. In contrast, the C S F level o f H V A at 3-7 y e a r s o f age w a s still h i g h e r t h a n that o f adults. T h e s e f i n d i n g s s u p p o r t e d t h o s e o f p r e v i o u s studies [14,34]. T h e p r e s e n t study d o c u m e n t e d a strong i n v e r s e correlation b e t w e e n age and C S F levels o f n e u r o t r a n s m i t t e r m e tabolites. T h e i m p o r t a n c e o f a g e - m a t c h e d c o n t r o l s in studies o f C S F n e u r o t r a n s m i t t e r m e t a b o l i t e s in the d e v e l o p i n g b r a i n is e m p h a s i z e d . T h e s e f i n d i n g s s u g g e s t v a r i a t i o n s in central n e u r o n a l activity or variations in the c l e a r a n c e o f t h e s e m e t a b o l i t e s f r o m the CSF. A l t h o u g h the d i f f e r e n c e s are statistically significant, the small n u m b e r o f p a t i e n t s and the fact that v a r i a b l e s (e.g., c i r c a d i a n c h a n g e s , sex, and dietary i n f l u e n c e s ) w e r e not c o n t r o l l e d , m a k e t h e s e findings preliminary. M o r e p a t i e n t s in e a c h age r a n g e n e e d to b e studied.
References [1] Berger PA, Faull KF, Kilkowski J, et al. CSF monoamine metabolites in depression and schizophrenia. Am J Psychiatry 1980;137: 174-80. [2] "l]raskman L, Asberg M, Bertilsson L, Sjostrand L. Monoamine metabolites in CSF and suicidal behavior. Arch Gen Psychiatry 1981; 38:631-6. [3] Anderson GM, Hoder EL, Shaywitz BA, Cohen DJ. Neurotransmitter precursors and metabolites in CSF of human neonates. Dev Med Child Neurol 1985;27:207-14. [4] Botez MI, Young SN, Bachevalier J, Gauthier S. Effects of folic acid and vitamin B12 deficiencies on 5-hydroxyindoleacetic acid in human cerebrospinal fluid. Ann Neurol 1982; 12:479-84.
3-Methoxy-4-hydroxyphenyl glycol Newborn Tryptophan Tyrosine
[5] Gerner RH, Fairbanks L, Anderson GM. Cerebrospinal fluid parameters in depressed, manic and schizophrenic patients compared to normal controls. Am J Psychiatry 1984; 14:1533-40. [6] Leekman JF, Cohen DJ, Shaywitz BA, Caparulo BK, Heninger GR, Bowers MB. CSF monoamine metabolites in child and adult psychiatric patients. Arch Gen Psychiatry 1980;37:677-81. [7] Young SN, Gauthier S, Anderson GM, et al. Tryptophan, 5-hydroxyindoleacetic acid and indoleacetic acid in human cerebrospinal fluid: Interrelationships and the influence of age, sex, epilepsy and anticonvulsant drugs. J Neurol Neurosurg Psychiatry 1980;43:438-45. [8] Curzon G, Kantamanehi BD, Bartlett JR, Bridges PK. Transmitter precursors and metabolites in human ventricular cerebrospinal fluid. J Neurochem 1976;26:613-5. [9] Gilman PK, Bartlett JR, Bridges PK, et al. Indolic substances in plasma, cerebrospinal fluid and frontal cortex of human subjects infused with saline or tryptophan. J Neurochem 1981;37:410-7. [10] Westerink BHC, Wirix E. On the significance of tyrosine for the synthesis and catabolism of dopamine in rat brain: Evaluation by HPLC with electrochemical detection. J Neurochem 1983;40:758-64. [11] Habel A, Yates CM, McQueen JK, Blackwood D, Ehon RA. Homovanillic acid and 5-hydroxyindoleacetic acid in lumbar cerebrospinal fluid in children with afebrile and febrile convulsions. Neurology 1981;31:488-91. [12] Shaywitz BA, Cohen D J, Bowers MB. Cerebrospinal fluid monoamine metabolites in neurological disorders of childhood. In: Wood JH, ed. Neurobiology of cerebrospinal fluid. New York: Plenum, 1980;219-36. [13] Seifert WE, Foxx JF, Butler IJ. Age effect on dopamine and serotonin metabolite levels in cerebrospinal fluid. Ann Neurol 1980;8: 38-42. [14] Langlais PJ, Walsh FX, Bird ED, Levy HL. Cerebrospinal fluid neurotransmitter metabolites in neurologically normal infants and children. Pediatrics 1985;75:580-6.
Yamamoto: CSF Neurotransmitters in Neonates 409
[15] Matson WR, Langlais PJ, Volicer L, Gamache PH, Bird E, Mark KA. n-Electrode three-dimensional liquid chromatography with electrochemical detection for the determination of neurotransmitters. Clin Chem 1984;30:1158-61. [16] Matson WR, Gamache PH, Beal MF, Bird E. EC array sensor concepts and data. Life Sci 1987;41:905-8. [17] Habel A, Yates CM, McQueen JK, Blackwood D, Elton RA. HomovaniUic acid and 5-hydroxyindoleacetic acid in lumbar cerebrospinal fluid in children with afebrile and febrile convulsions. Neurology 1981 ;31:488-91. [18] Silverstein FS, Donn S, Buchanan K, Johnston MJ. Concentrations of homovanillic acid and 5-hydroxyindoleacetic acid in cerebrospinal fluid from human infants in the perinatal period. J Neurochem 1984;43:1769-72. [19] Rogers KJ, Dubowitz V. 5-Hydroxyindoles in hydrocephalus. A comparative study of cerebrospinal fluid and blood levels. Dev Med Child Neurol 1970;12:461-6. [20] Wirz-Justice A, Wehr T. Seasonality in biochemical determinations: A source of variance and a clue to the temporal incidence of affective illness. Psychiatr Res 1979; 1:53-8. [21] Brewerton TD, Berrettini WH, Nurnberger Jl, Linnoila M. Analysis of seasonal fluctuations of CSF monoamine metabolites and neuropeptides in normal controls: Findings with 5-HIAA and HVA, Psychiatr Res 1988;23:257-65. [22] Swedo SE, Kruesi MJP, Leonard HL, et al. Lack of seasonal variation in pediatric lumbar cerehrospinal fluid neurotransmitter metabolite concentration. Acta Psychiatr Scand 1989;80:644-9. [231 Banki CM, Molnar G. Cerebrospinal fluid 5-hydroxyindoleacetic acid as an index of central serotonergic processes. Psychiatr Res 1981;5:23-32. [241 Anderson O, Johansson B, Svennerholm C. Monoamine metabolites in successive samples of spinal fluid. Acta Neurol Scand 1981 ;63:247-54.
410
PEDIATRIC NEUROLOGY
Vol. 7 No. 6
[25] Bertilsson L, Asberg M, Lantto O. Gradient of monoamine metabolites and cortisol in cerebrospinal fluid of psychiatric patients and healthy controls. Psychiatr Res 1982;6:77-83. [26] Kruesi MJP, Swedo SE, Hamburger SD, Potter WZ, Rapoport JL. Concentration gradient of monoamine metabolites in children and adolescents. Biol Psychiatry 1988:24:507-14. [27] Hender T, Lundborg P. Serotoninergic development in the postnatal rat brain. J Neural Transm 1980:49:257-79. [28] Volicer L, Direnfeld LK, Freedman M, Albert ML, Langlais PJ, Bird ED. Serotonin and 5-hydroxyindoleacetic acid in CSF: Difference in Parkinson's disease and dementia of the Alzheimer's type. Arch Neurol 1985;42:127-9. [291 Volicer L, Langlais PJ, Matson WR, Mark KA, Gamache PH. Serotoninergic system in dementia of the Alzheimer type: Abnormal forms of 5-hydroxytryptophan and serotonin in cerebrospinal fluid. Arch Neurol 1985:42:1158-61. [30] LinnoUa M, Jacobson KA, Marshall TH, Miller TL, Kirk KL. Liquid chromatographic assay tor cerebrospinal fluid serotonin. Life Sci 1986;38:687-94. [31] Stone TW, Connick JK. Quinolinic acid and other kynurenines in the central nervous system. Neuroscience 1985; 15:597-617. [32] Freese A, Swartz KJ, During MJ. Martin JB. Kynurenine metabolites of tryptophan: Implications for neurologic diseases. Neurology 1990:40:691-5. [33] Gal EM, Sherman AD. L-Kynurenine: Its synthesis and possible regulatory function in brain. Neurochem Res 1980;5:223-39. [341 Hender J, Lundell KH, Beese GR, Mueller RA, Hender T. Developmental variations in CSF monoamine metabolites during childhood. Biol Neonate 1986;49:190-7.
Changes in CSF Neurotransmitters During the First Year of Life Hitoshi Yamamoto, MD
Changes in cerebrospinal fluid neurotransmitter metabolites in 25 children younger than 1 year of age were analyzed to assess maturation of the central nervous system and were compared to cerebrospinal fluid from older children and adults. Significant inverse correlations (P < .05) with aging were observed for tryptophan, 5-hydroxytryptophan, 5-hydroxyindoleacetic acid, kynurenine, tyrosine, dopa, dopamine, dihydroxyphenylacetic acid, homovanUlic acid, norepinephrine, and 3-methoxy-4-hydroxyphenyiglycol concentrations. There were no significant differences observed with respect to age in the cerebrospinal fluid serotonin, 3-hydroxykynurenine, and 3-methoxytyramine concentrations. This study suggests that changes in the major cerebrospinal fluid neurotransmitters occur with increasing age during the neonatal period. Because these findings are preliminary, additional patients require study. Yamamoto H. Changes in C S F neurotransmitters during the first year of life. Pediatr Neurol 1991;7:406-10.
Introduction The first year of life is a period of very rapid development and is of special interest in measuring cerebrospinal fluid (CSF) levels of neurotransmitter metabolites. CSF levels of neurotransmitter metabolites have been widely studied to ascertain neurotransmitter turnover in the central nervous system (CNS). Problems remain regarding the degree to which the peripheral levels are correlated with brain levels and whether alterations occur in particular neuroiogic disorders [1-3]. The fact that medications effective in ameliorating symptoms in some neurologic disorders have specific effects on CSF neurotransmitters, as well as altered CSF neurotransmitter metabolite levels in different diagnostic groups, has encouraged continued investigation of neurotransmitters in neurologic disorders of childhood. C S F concentrations of h o m o v a n i l l i c acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA) have
From the Department of Pediatrics; St. Marianna University School of Medicine; Kawasaki 213, Japan.
406 PEDIATRIC NEUROLOGY Vol. 7 No. 6
been determined in adult controls and in adults with neurologic disorders [2,4,5]. In adult neurologic disorders, an age effect has been reported for CSF HVA and possibly for CSF 5-HIAA [6,7]. The availability of the precursor amino acids in brain has been believed to affect central serotonin synthesis [8,9] and, to a lesser extent, dopamine and norepinephrine synthesis 110]. The issue of precursor control of neurotransmitter synthesis in children can be studied by examining correlations between precursors and metabolites. The ontogenetic pattern of precursor levels is also of interest, regardless of the extent to which these levels influence the rate of neurotransmitter synthesis. The usefulness o f C S F neurotransmitter metabolite measures in infants has been limited by a lack of normal control data. In disorders believed not to disturb CNS neurotransmitter metabolism, CSF obtained by diagnostic lumbar puncture has been analyzed for neurotransmitters [3,11]. Although an age effect for HVA and possibly 5H I A A has been observed in some of these studies [6,12], mean values for discrete age intervals have been presented in only a few reports [ 13,141. Individual variability of C S F neurotransmitter metabolites, including kynurenine (KYN) and 3-hydroxykynurenine (3-OHKYN), in neonates has not been studied. In the present study, the levels of tryptophan (TRP), 5-hydroxytryptophan (5-HTP), serotonin (5-HT), 5-HIAA, KYN, 3-OHKYN, tyrosine (TYR), L-dihydroxyphenylalanine (L-DOPA), d o p a m i n e (DA), dihydroxyphenylacetic acid (DOPAC), HVA, norepinephrine (NE), 3-methoxy-4-hydroxyphenyl glycol (MHPG), and 3-methoxytyramine (3MT) were measured in the CSF of neurologically normal infants, children, and adults in order to establish normal values as well as to characterize interrelationships among these compounds. The tryptophan and tyrosine metabolic pathways are illustrated in Figure 1.
Methods CSF was obtained from 25 neurologically normal infants (newborn to 11 months), 7 children (3-7 years), and 9 adults. Samples were obtained from infants and children who required lumbar puncture for diagnosis of possible meningitis and from adults with headache or
Communications should be addressed to: Dr. Yamamoto; Neurology Service (V-151); VA Medical Center; 3350 La Jolla Village Drive; San Diego, CA 92161. Received March 7, 1991; accepted July 1, 1991.
/, TRP
/-
KYN
Each of the compounds was analyzed for the 5 age groups using a standard computerized analysis of variance (ANOVA) statistical procedure (Statistical Analysis System).
) 3-OHKY
)5-HTP
) 5 - H T --~-~ 5 - H I A A
Results
3-MT TYR
) L-DOPA
DOPAC
)DA '~NE
)HVA
--~-~ MHPG
Figure 1. The tryptophan and tyrosine metabolic pathways. Abbreviations: TRP, tryptophan; 5-HTP, 5-hydroxytryptophan; 5-HT, serotonin; 5-HIAA, 5-hydroxyindoleacetic acid; KYN, kynurenine; 3-OHKYN, 3hydroxykynurenine; TYR, tyrosine; L-DOPA, L-dihydro.ryphenylalanine; DA, dopamine; DOPAC, dihydroxyphenylacetic acid; HVA, homovanillic acid; NE, norepinephrine; MHPG, 3-methoxy-4-hydroxyphenyl glycol; 3-MT, methoxytyramine. neuropathy; the adults and children were later shown to be neurologically normal. The infants' medical histories were examined carefully and none had any of the following signs: prematurity, seizures, asphyxia, hydrocephalus, sepsis, intraventricular hemorrhage, or CNS medications. CSF was collected by lumbar puncture while the patient was in the lateral decubitus position and was immediately frozen at -70°C for later analysis. The total volume of CSF removed from each infant or child was typically 3 ml; 3-5 ml samples were taken from adult patients. CSF was collected in 2 aliquots (from infants and children) or 3 aliquots (from adults). The last aliquot (0.5 ml) was always used for analysis. Analysis of neurotransmitter metabolites was accomplished by direct injection of CSF (50 lad onto a computer-controlled, high-performance, liquid chromatographic (HPLC) analyzer. This instrument, the Neurochem neurochemical analyzer (ESA, Bedford, MA), consists of a gradient HPLC system and 16 high-sensitivity coulometric electrochemical detectors. The concept and inherent advantages of this multielectrode HPLC system have been described elsewhere [15,16]. The gradient profile, column, and detector potentials used in this study were similar to those previously described [16]. The mobile phase solutions were obtained directly from ESA (Bedford, MA) and had the following compositions: Mobile phase A - 0 . 1 sodium phosphate with 10 mg/L sodium dodecyl sulfate at pH 3.35; Mobile phase B - 5 0 % methanol/water with 50 mg/L sodium dodecyl sulfate at pH 3.45. All standards were obtained from Sigma Chemical Company (St. Louis, MO). Table 1.
The patients were placed into 5 age groups for analysis: Group 1 : 5 newborns younger than 1 month of age; Group 2 : 1 0 infants 1-5 months (mean age: 3.7 mos); Group 3 : 1 0 infants 6-11 months (mean age: 8.1 mos); Group 4 : 7 children 3-7 years; and, Group 5 : 9 adults. The variations in the individual CSF neurotransmitter metabolite concentrations with age are listed in Tables 1 and 2. The levels of CSF TRP, TYR, L-DOPA, and 5-HIAA were highest in group 1 and decreased significantly with increasing age (P < .05). No significant difference was observed between the levels in groups 4 and 5. The levels of CSF 5-HTP were highest in group 1 and decreased significantly with increasing age (P < .05); however, there were no significant differences between the values in groups 2 and 3 or in groups 4 and 5. The levels of CSF KYN and DA in infants younger than 1 year of age were higher than those of the older children and after 1 year of age approached the adult levels. The concentrations of CSF MHPG, DOPAC, and NE decreased significantly in group 3 (P < .05) and were reduced to near adult levels after 6 months of age. CSF HVA concentrations also decreased significantly (P < .05) during the first year of life and were still above adult levels in group 4. The levels of CSF 5-HT, 3-OHKYN, and 3MT did not differ significantly with increasing age. Discussion
Concentrations of neurotransmitter metabolites in CSF are determined by the rates of synthesis, release, and degradation of the parent compounds as well as the rate and efficiency of elimination of metabolites from brain
Changes in CSF TRP metabolites with age* Group 1 Newborns (N = 5)
Group 2 1-5 mos (N = 10)
Group 3 6-11 mos (N : 10)
882.9 _+ 101.6
739.2 + 68.0 +
610.4 _+ 105.6 ~
337.6 _+55.2 §
289.1 _+35.2
5-Hydroxytryptophan
2.7 + 0.2
1.9 + 0.4 +
1.9 _+0.5
1.2 + 0.3 §
1.1 _+0.3
Serotonin
.04 + .02
.04 + .01
.05 _+.02
.04 + .02
.05 + .02
5-Hydroxyindoleacetic acid
130.4 + 6.9
81.0 + 11.1'
48.2 _+6.6 :~
20.3 _+5.4 §
18.3 + 7.3
Kynurenine
41.8 _+7.0
38.6 -+ 9.9
37.9 + 8.3
12.5 + 4.6§
11.7 + 7.2
3-Hydroxykynurenine
0.15 + .04
0.12 -+ .03
0.14 _+.04
0.13 + .03
0.14 _+.07
Compound Tryptophan
Group 4 3-7 yrs (N = 7)
Group 5 Adults (N = 9)
* All data expressed as ng/ml CSF (mean + S.D.). + Significantly different from Group 1, P < .05. Significantly different from Group 2, P < .05. § Significantly different from Group 3, P < .05.
Yamamoto: CSF Neurotransmitters in Neonates
407
Table 2.
Changes in CSF tyrosine metabolites with age*
Compound
Tyrosine
Group 1 Newborn (N = 5)
Group 2 1-5 mos (N : 10)
Group 3 6-11 mos (N = 10)
Group 4 3-7 yrs (N : 7)
Group 5 Adults (N = 9)
5,525.3 _+1,114
4,787 +_1,589
4.190.3+ 1,254.3
2.201 _+770§
1.877 + 203
L-Dihydroxyphenylalanine
0.76 -+0.32
0.54 i 0.14+
0.35 _+0.12:~
0.18 _+.05§
0.2 _+0.1
Dopamine
0.43 + 0.11
0.49 _+0.16
(I.4 + .07
0.15 + .04~
0.18 + .05
Dihydroxyphenylaceticacid
0.71 _+0.15
0.51 _+0.21
0.36 -+0.14'
(t.35 +_0. t2
0.29 _+0.13
Homovanillicacid
156.9 + 16.7
100.2 + 18.7+
65.7 + 13:~
63.9 _+17.1
33.6 + 11.21[
Norepinephrine
0.48 + 0.12
(I.47 + 0.2
0.31 _+0.11;
0.3 _+.07
0.31 _+0.14
3-Methoxy-4-hydroxyphenyl glycol
22.8 + 3.1
21.9 + 3.2
13.5 + 2.6+
10.7 + 2.9
7.9 -+ 1.9
3-Methoxytyramine
0.21 _+.(12
0.2 _+.02
0.2 _+.03
(I.19 _+.01
0.18 _+.02
* All data expressed as ng/ml CSF (mean i S.D.). + Significantlydifferentfrom Group 1, P < .05. Significantlydifferentfrom Group 2, P < .05. § Significantlydifferentfrom Group 3, P < .05. ¶ Significantlydifferentfrom Group 4, P < .05. and CSE Maturational changes in storage pools, rates of tumover, transport systems, intra- and extra-neuronal metabolism, rates of CSF production, and brain and spinal column morphology all may alter CSF metabolite concentrations. In previous studies, several investigators reported that the CSF levels of TRE 5-HIAA, TYR, HVA, and MHPG in neonates and the values for neonates were considerably higher than those of older children and adults. Approximately 50% of patients in each age group had fever, but fever has been demonstrated not to affect 5-HIAA and HVA levels [17]. The compound concentrations reported here in the newborn to 1l-month-old infants are in agreement with previous reports of neurologically normal infants (Table 3); however, there is little normal control data for CSF neurotransmitter metabolites and CSF levels of KYN and 3-OHKYN have not been previously reported in neonates. Seasonal variations of CSF neurotransmitter metabolite concentration have been reported in several adult psychiatric patients and normal individuals [20,21]; however, Swedo et al. reported that there were no seasonal differences in CSF concentrations of MHPG, HVA, or 5-HIAA in a pediatric population [22]. In the present study, the time of year of the lumbar puncture varied and the possibility of seasonal variations of CSF neurotransmitter metabolite levels in neonates cannot be determined. The concentration gradient of CSF also may affect the results. A lumbar-cistemal concentration gradient exists for HVA and 5-HIAA in adults [23-25]; therefore, the concentration may depend on the volume of the sample obtained. In children, a concentration gradient of CSF monoamine metabolites has also been reported [26]. It is evident that for the purpose of comparison, samples from children and adults must be collected in precisely the same manner and metabolites must be measured in the same fraction of the collected fluid.
408 PEDIATRICNEUROLOGY Vol.7 No. 6
The levels of CSF TRP and its metabolites 5-HTP and 5-HIAA were highest in newborns (younger than 1 month) and decreased gradually with age. It is probable that the age-related changes in CSF neurotransmitter metabolites are related to differences in neurotransmitter turnover rate and the rate and efficiency of elimination from CSE The mode of metabolic clearance is changed from the newbom stage through adulthood. It is reported that the main mode of clearance is via the CSF to the choroid plexus in newboms, whereas in adults these compounds exit directly from the brain to the blood, probably across the bloodbrain barrier [27]. In the present study, the levels of CSF 5-HT did not differ with increasing age and were lower than the levels reported in previous studies [28-30]. This difference may be due to the different method of detection in this study or by the volume of CSF obtained [25,26]. The presence of 5-HT in CSF may be related t o diffusion of unmetabolized 5-HT which has been released from nerve terminals. Levels of 5-HT also depend on the degree of serotoninergic activity in the CNS. The CSF levels of KYN and 3-OHKYN in neonates have not been reported. These are metabolites of TRP that are excitotoxic and are implicated in the pathogenesis of certain human neurodegenerative and seizure disorders [31,32]. In the present study, the levels of CSF KYN in neonates were higher than those of children and adults. In contrast, CSF levels of 3-OHKYN did not change with increasing age. According to Gal and Sherman, in rats the metabolism of TRP via the KYN pathway is about 45% of the metabolism of TRP to 5-HT [33]; this percentage may change with age. Except for 3MT, a significant inverse correlation was observed between age and the CSF concentration of TYR metabolites. The c o n c e n t r a t i o n of M H P G , NE, and DOPAC decreased rapidly and reached adult levels by 6-11 months of age. The L-DOPA level decreased earlier
Table 3. Literature summary of reports of CSF neurochemical levels in infants
Authors
Patients
Assay
Age (mos)
N
TRP (ng/ml)
5-HIAA (ng/ml)
TYR (ng/ml)
HVA (ng/ml)
MHPG (ng/ml) 23 ± 3 22 ± 3 16 ± 3
Present patients
Normal infants
HPLCEC
0-1 1-5 6-11
5 10 10
833 ± 102" 739 ± 68 610 ± 106
130 ± 7 81 ± 11 48+7
5,525 ± 1,114 4,788 ± 1,590 4,190 ± 1,254
157 ± 17 100 ± 19 66 ± 13
Anderson et al. [3]
Normal infants
HPLCEC/F
0-1
17
1,250 ± 100
143 ± 9
7,940 ± 1,480
184 ± 15
Langlais et al. [14]
Normal infants
HPLCEC
0-1 4-7 8-12
12 8 5
143 ±40 62 ± 16 58 ± 24
197 ± 62 148 ± 48 157 ±44
Silverstein et al. [18]
Suspicion of infection
HPLCEC
NB 0-6
19 13
104±7 85±7
129 ± 11 159 ± 9
Seifert et al. [13]
Neurologic disorder
GC-MS
0-12
14
-43**
-120"*
Rogers and Dubowitz [ 19]
Miscellaneous disorders
Fluorometry
0-12
13
99± 11
29± 14 15 ± 3 13 ± 2
* Values are mean + S.D. ** Data were not reported in numerical form (approximated). Abbreviations: 5-HIAA = 5-Hydroxyindoleacetic acid EC/F = Electrochemical detection/fluorometry GC-MS = Gas chromatography-mass spectrometry HVA = Homovanillic acid
MHPG NB TRP TYR
= = = =
in n e o n a t a l life. In contrast, the C S F level o f H V A at 3-7 y e a r s o f age w a s still h i g h e r t h a n that o f adults. T h e s e f i n d i n g s s u p p o r t e d t h o s e o f p r e v i o u s studies [14,34]. T h e p r e s e n t study d o c u m e n t e d a strong i n v e r s e correlation b e t w e e n age and C S F levels o f n e u r o t r a n s m i t t e r m e tabolites. T h e i m p o r t a n c e o f a g e - m a t c h e d c o n t r o l s in studies o f C S F n e u r o t r a n s m i t t e r m e t a b o l i t e s in the d e v e l o p i n g b r a i n is e m p h a s i z e d . T h e s e f i n d i n g s s u g g e s t v a r i a t i o n s in central n e u r o n a l activity or variations in the c l e a r a n c e o f t h e s e m e t a b o l i t e s f r o m the CSF. A l t h o u g h the d i f f e r e n c e s are statistically significant, the small n u m b e r o f p a t i e n t s and the fact that v a r i a b l e s (e.g., c i r c a d i a n c h a n g e s , sex, and dietary i n f l u e n c e s ) w e r e not c o n t r o l l e d , m a k e t h e s e findings preliminary. M o r e p a t i e n t s in e a c h age r a n g e n e e d to b e studied.
References [1] Berger PA, Faull KF, Kilkowski J, et al. CSF monoamine metabolites in depression and schizophrenia. Am J Psychiatry 1980;137: 174-80. [2] "l]raskman L, Asberg M, Bertilsson L, Sjostrand L. Monoamine metabolites in CSF and suicidal behavior. Arch Gen Psychiatry 1981; 38:631-6. [3] Anderson GM, Hoder EL, Shaywitz BA, Cohen DJ. Neurotransmitter precursors and metabolites in CSF of human neonates. Dev Med Child Neurol 1985;27:207-14. [4] Botez MI, Young SN, Bachevalier J, Gauthier S. Effects of folic acid and vitamin B12 deficiencies on 5-hydroxyindoleacetic acid in human cerebrospinal fluid. Ann Neurol 1982; 12:479-84.
3-Methoxy-4-hydroxyphenyl glycol Newborn Tryptophan Tyrosine
[5] Gerner RH, Fairbanks L, Anderson GM. Cerebrospinal fluid parameters in depressed, manic and schizophrenic patients compared to normal controls. Am J Psychiatry 1984; 14:1533-40. [6] Leekman JF, Cohen DJ, Shaywitz BA, Caparulo BK, Heninger GR, Bowers MB. CSF monoamine metabolites in child and adult psychiatric patients. Arch Gen Psychiatry 1980;37:677-81. [7] Young SN, Gauthier S, Anderson GM, et al. Tryptophan, 5-hydroxyindoleacetic acid and indoleacetic acid in human cerebrospinal fluid: Interrelationships and the influence of age, sex, epilepsy and anticonvulsant drugs. J Neurol Neurosurg Psychiatry 1980;43:438-45. [8] Curzon G, Kantamanehi BD, Bartlett JR, Bridges PK. Transmitter precursors and metabolites in human ventricular cerebrospinal fluid. J Neurochem 1976;26:613-5. [9] Gilman PK, Bartlett JR, Bridges PK, et al. Indolic substances in plasma, cerebrospinal fluid and frontal cortex of human subjects infused with saline or tryptophan. J Neurochem 1981;37:410-7. [10] Westerink BHC, Wirix E. On the significance of tyrosine for the synthesis and catabolism of dopamine in rat brain: Evaluation by HPLC with electrochemical detection. J Neurochem 1983;40:758-64. [11] Habel A, Yates CM, McQueen JK, Blackwood D, Ehon RA. Homovanillic acid and 5-hydroxyindoleacetic acid in lumbar cerebrospinal fluid in children with afebrile and febrile convulsions. Neurology 1981;31:488-91. [12] Shaywitz BA, Cohen D J, Bowers MB. Cerebrospinal fluid monoamine metabolites in neurological disorders of childhood. In: Wood JH, ed. Neurobiology of cerebrospinal fluid. New York: Plenum, 1980;219-36. [13] Seifert WE, Foxx JF, Butler IJ. Age effect on dopamine and serotonin metabolite levels in cerebrospinal fluid. Ann Neurol 1980;8: 38-42. [14] Langlais PJ, Walsh FX, Bird ED, Levy HL. Cerebrospinal fluid neurotransmitter metabolites in neurologically normal infants and children. Pediatrics 1985;75:580-6.
Yamamoto: CSF Neurotransmitters in Neonates 409
[15] Matson WR, Langlais PJ, Volicer L, Gamache PH, Bird E, Mark KA. n-Electrode three-dimensional liquid chromatography with electrochemical detection for the determination of neurotransmitters. Clin Chem 1984;30:1158-61. [16] Matson WR, Gamache PH, Beal MF, Bird E. EC array sensor concepts and data. Life Sci 1987;41:905-8. [17] Habel A, Yates CM, McQueen JK, Blackwood D, Elton RA. HomovaniUic acid and 5-hydroxyindoleacetic acid in lumbar cerebrospinal fluid in children with afebrile and febrile convulsions. Neurology 1981 ;31:488-91. [18] Silverstein FS, Donn S, Buchanan K, Johnston MJ. Concentrations of homovanillic acid and 5-hydroxyindoleacetic acid in cerebrospinal fluid from human infants in the perinatal period. J Neurochem 1984;43:1769-72. [19] Rogers KJ, Dubowitz V. 5-Hydroxyindoles in hydrocephalus. A comparative study of cerebrospinal fluid and blood levels. Dev Med Child Neurol 1970;12:461-6. [20] Wirz-Justice A, Wehr T. Seasonality in biochemical determinations: A source of variance and a clue to the temporal incidence of affective illness. Psychiatr Res 1979; 1:53-8. [21] Brewerton TD, Berrettini WH, Nurnberger Jl, Linnoila M. Analysis of seasonal fluctuations of CSF monoamine metabolites and neuropeptides in normal controls: Findings with 5-HIAA and HVA, Psychiatr Res 1988;23:257-65. [22] Swedo SE, Kruesi MJP, Leonard HL, et al. Lack of seasonal variation in pediatric lumbar cerehrospinal fluid neurotransmitter metabolite concentration. Acta Psychiatr Scand 1989;80:644-9. [231 Banki CM, Molnar G. Cerebrospinal fluid 5-hydroxyindoleacetic acid as an index of central serotonergic processes. Psychiatr Res 1981;5:23-32. [241 Anderson O, Johansson B, Svennerholm C. Monoamine metabolites in successive samples of spinal fluid. Acta Neurol Scand 1981 ;63:247-54.
410
PEDIATRIC NEUROLOGY
Vol. 7 No. 6
[25] Bertilsson L, Asberg M, Lantto O. Gradient of monoamine metabolites and cortisol in cerebrospinal fluid of psychiatric patients and healthy controls. Psychiatr Res 1982;6:77-83. [26] Kruesi MJP, Swedo SE, Hamburger SD, Potter WZ, Rapoport JL. Concentration gradient of monoamine metabolites in children and adolescents. Biol Psychiatry 1988:24:507-14. [27] Hender T, Lundborg P. Serotoninergic development in the postnatal rat brain. J Neural Transm 1980:49:257-79. [28] Volicer L, Direnfeld LK, Freedman M, Albert ML, Langlais PJ, Bird ED. Serotonin and 5-hydroxyindoleacetic acid in CSF: Difference in Parkinson's disease and dementia of the Alzheimer's type. Arch Neurol 1985;42:127-9. [291 Volicer L, Langlais PJ, Matson WR, Mark KA, Gamache PH. Serotoninergic system in dementia of the Alzheimer type: Abnormal forms of 5-hydroxytryptophan and serotonin in cerebrospinal fluid. Arch Neurol 1985:42:1158-61. [30] LinnoUa M, Jacobson KA, Marshall TH, Miller TL, Kirk KL. Liquid chromatographic assay tor cerebrospinal fluid serotonin. Life Sci 1986;38:687-94. [31] Stone TW, Connick JK. Quinolinic acid and other kynurenines in the central nervous system. Neuroscience 1985; 15:597-617. [32] Freese A, Swartz KJ, During MJ. Martin JB. Kynurenine metabolites of tryptophan: Implications for neurologic diseases. Neurology 1990:40:691-5. [33] Gal EM, Sherman AD. L-Kynurenine: Its synthesis and possible regulatory function in brain. Neurochem Res 1980;5:223-39. [341 Hender J, Lundell KH, Beese GR, Mueller RA, Hender T. Developmental variations in CSF monoamine metabolites during childhood. Biol Neonate 1986;49:190-7.
