вход по аккаунту


Cerebrospinal fluid IgG in childhood The establishment of reference values.

код для вставкиСкачать
This work was supported by a Grant-in-Aid for Special Project
Research of Selected Intractable Neurological Disorders (No.
61109006) from the Ministry of Education, Science, and Culture,
Cerebrospinal Fluid IgG
in Childhood: The
Establishment of
Reference Values
We thank the staffs of the Departments of Neurology and Pathology
of Biwako Geriatric Hospital for preparing the autopsy samples.
Robert S. Rust Jr, MD,*tf W. Edwin Dodson, MD,”t
and John L.Trotter, M D t
pathophysiological role of benzodiazepine receptors in
1. Reisine RD, Yamamura HI, Bird ED, et al. Pre- and postsynaptic neurochemical alterations in Alzheimer’s disease. Brain
Res 1978;159:447-45 1
2. Cross AJ, Crow TJ, Ferrier IN, et al. Serotonin receptor
changes in dementia of the Alzheimer type. J Neurochem
3. Bed MF, Mazurek MF, Tran VT,et al. Reduced numbers of
somatostatin receptors in the cerebral cortex in Alzheimer’s
disease. Science 1985;229:289-291
4. Shimohama S, Taniguchi T, Fujiwara M, Kameyama M.
Changes in nicotinic and muscarinic cholinergic receptors in
Alzheimer-type dementia. J Neurochem 1986;46:288-293
5. Shimohama S, Taniguchi T, Fujiwara M, Kameyama M.
Biochemical characterization of a-adrenergic receptors in human brain and changes in Alzheimer-type dementia J Neurochem 1986;47:1294-1 301
6. Shimohama S, TanigUchi T, Fujiwara M, Kameyama M.
Changes in P-adrenergic receptor subtypes in Alzheimer-type
dementia. J Neurochem 1987;48:12 15-122 1
7. Tallman JF, Paul SM, Skolnick P, Gallager DW. Receptors for
the age of anxiety: pharmacology of the benzodiazepines. Science 1980;207:274-28 I
8. Skolnick P, Paul SM. Benzodiazepine receptors in the central
nervous system. Int Rev Neurobiol 1982;23:103-140
9. Owen F, Poulter M, Waddington J, Mashal RD. 3H-R05-4864
and ’H-flunitrazepam binding in kainate-lesioned rat striatum
and in temporal cortex of brains from patients with senile dementia of the Alzheimer type. Brain Res 1983;278:373-375
10. Cross AJ, Crow TJ, Jonson JA, et al. Studies on neurotransmitter receptor systems in neocortex and hippocampus in senile
dementia of the Alzheimer-type. J Neurol Sci 1984;64:109117
11. Shimohama S, Taniguchi T, Fujiwara M, Kameyama M.
Biochemical characterization of the nicotinic cholinergic receptors in human brain: binding of (-)-3H-nicotine. J Neurochem
12. Bosmann HB, Penney DP, Case KR, et al. Diazepam receptors:
specific binding of 3H-diazepam and 3H-flunitrazepam to rat
brain subfractions. FEBS Lett 1978;87:199-202
13. Young WS, Kuhar MJ. Autoradiographic localisation of benzodiazepine in the brains of humans and animals. Nature
14. Ellison DW, Bed MF, Mazurek MF, et al. A postmortem study
of amino acid neurotransmitters in Alzheimer’s disease. Ann
Neurol 1986;20:616-62 1
15. Perry TL, Young VW, Bergeron C, et al. Amino acids,
glutathione, and glutathione transferase activity in the brains of
patients with Alzheimer’s disease. Ann Neurol 1987;21:331336
16. Nielsen M, Braestrup C, Squires R. Evidence for a late evolutionary appearance of brain-specific benzodiazepine receptors:
an investigation of 18 vertebrates and 5 invertebrate species.
Brain Res 1978;141:342-346
A retrospective analysis of quantitative and qualitative
immunoglobulin G ( 1 6 ) results from 253 children who
were either medically and neurologically normal or
highly unlikely to have abnormalities of CSF IgG is reported. Normal values in this reference population vary
with age for CSFlalbumin IgG ratio and CSFlserum IgG
index and are significantly different from the adult reference values. The rate of false positivity is lower for
quantitative values than for qualitative IgG determinations (oligoclonal bands).
Rust RS Jr, Dodson WE, Trotter JL.
Cerebrospinal fluid IgG in childhood:
the establishment of reference values.
Ann Neurol 1988;23:406-410
Despite accumulating evidence of significant developmental changes in the blood-brain barrier that directly
affect CSF globulin concentrations El, 21, comparatively few studies have attempted to characterize CSF
immunoglobulin G ( 1 6 ) in reference populations of
children [ l , 3-16]. Thus childhood values are usually
compared to reference ranges established for adults.
We reviewed the results of all CSF IgG evaluations in
children at the St Louis Children’s Hospital between
1976 and 1987 to designate a reference group for determination of reference childhood ranges and rates of
false positivity for various methods of IgG determination and to examine possible effects of age or sex on
CSF IgG concentrations.
Materials and Methods
CSF cell counts, protein, and glucose were determined by
standard methodology. CSF and serum specimens were refrigerated at 4°C for periods of 6 to 60 hours (usually less
than 24 hr) prior to analysis. Quantitative determination of
CSF and serum IgG and albumin (Alb) was performed by
From the Departments of *Pediatrics and ?Neurology and Neurological Surgery, and the SMcDonnell Center for Studies of Higher
Brain Function, Washington University School of Medicine, St
Louis, MO 63110.
Received Jul 13, 1987, and in revised form Aug 18 and Oct 12.
Accepted for publication Oct 12, 1987.
Address correspondence to Dr Trotter, Washington University
School of Medicine, Department of Neurology, Box 8111, 660
South Euclid Ave, St Louis, MO 63 110.
406 Copyright 0 1988 by the American Neurological Association
"rocket" electroimmunodiffusion by the method of Towellotte rl7, 181 or by immunonephelometry. Concentrations
of IgG and Alb determined by either of these methods were
employed to calculate the CSF IgGiAlb ratio (IgG,,~/AIb,,~)
[171, and the CSF/serum IgG index ([IgGcs&&m}
119, 201. Qualitative CSF IgG abnormalities were assessed by agarose gel electrophoresis (AGE)
and/or isoelectric focusing on polyacrylamide gel (IEF) for
homogeneous bands in the gamma globulin region {18).
Midcathodic and end-cathodic CSF bands, found in the CSF
of normal individuals [20}, and alkaline CSF bands corresponding to serum bands of equal or greater intensity at a
given isoelectric point were excluded.
Medical records of all children (n = 640) in whom lumbar
CSF IgG (n = 701) and simultaneous serum IgG (n = 528)
values were determined were comprehensively reviewed to
exclude data from patients with degenerative, demyelinative,
neoplastic, infectious, or inflammatory disorders; acute encephalopathy; complicated or poorly controlled epilepsy;
movement disorders; neuropathy; hydrocephalus; acute
stroke; CNS hemorrhage; or therapeutic immunosuppression. Additional data were excluded because CSF Alb concentration was unmeasurable by electroimmunodiffusion (n
= 5) or because CSF contained 50 red blood celldmm3 (n =
17). The remaining children (n = 253) comprised our reference population for CSF IgG/Alb ratio (n = 255), CSF/
serum IgG index (n = 193), CSF AGE (n = 177), and CSF
IEF (n = 85). Average age was 8.8 2 5.6 years (range 0.120.5). Diagnoses included static encephalopathy (35%),
well-controlled idiopathic epilepsy (20%), migraine (13943,
developmental abnormalities of the CNS (6%),psychiatric
disease (5.5%), vertigo/syncope (3.9%), pseudotumor cerebri (3%), concussion (1.9%), remote cerebrovascular disease
(1.9%), aminoaciduria (l.6%>,muscle disease (1.2%), or
learning disorders (1.2%); 1.9% had other conditions; and
3.9% had no neurological disease. Overall, more than 60%
of patients were normal on neurological examination and had
self-limited neurological complaints that were quite unlikely
to involve abnormalities of CSF, serum immunoglobulins, or
blood-brain barrier function.
Statistical evaluation included regression analysis of CSF
IgG/Alb ratio and CSF/serum IgG index plotted against age
for each sex and for the entire normal population. Linear
regression analyses were also performed for discrete 6month blocks during the first 2 years of life, and for the
intervals 1.5 to 8 and 8 to 20.5 years of age. Statistical
significance was determined by evaluating the t statistic calculated for each correlation coefficient. For discrete comparisons of subpopulations by age and sex, a two-tailed Student's
t test was applied. Significance was assigned when p 5 0.05.
Data are presented as mean values 2 the standard deviation.
The upper normal limits were defined as the mean + 2
standard deviations of the mean.
There were no significant differences between males (n
= 141) and females (n = 114) in age, CSF IgG ratio,
or CSF/serum IgG index; thus these data were combined for further analysis. Because mean IgG and Alb
values by electroimmunodiffusion (n = 182) and im-
Ratio vs. Age
I .
. . .. .
. .
Fig 1 . Scattergram of CSF IgGIAlb ratio as a function of age
for reference patients less than 2 years old (A) and for the entire
reference popukation (B). Solid lines indicate 95% confidence
limits (mean ? 2 standard deviations) and hatched lines indicate mean values.
munonephelometry (n = 7 3 ) did not differ significantly, these data were also combined. The CSF IgG
ratio varied significantly with age for the population.
The mean ratio value for children less than 1.5 years of
age (0.097 ? 0.038, n = 39) was lower than the mean
for children greater than 1.5 years of age (0.128 &
0.043, n = 216, p 5 0.001). The upper normal limits
for CSF IgG/Alb ratio values were 0.17 for children
less than 1.5 years old and 0.22 for older children. The
distribution of the data points for ratio is shown in
Figure 1. There was a weak correlation between ratio
and age overall by regression analysis (r = 0.27, p <
0.001, n = 255), but subgroup analysis of individual
age epochs demonstrated that most of this variation
was accounted for by children 0.5 to 1.5 years of age ( r
= 0.654,p 5 0.0001, n = 34).
There were also age-related changes in the CSF/
serum IgG index (Fig 2). Children younger than 1.5
years old had a higher mean index (0.6 +- 0.21, n =
23) than older children (0.48 & 0.15, n = 170, p 5
0.0001). The upper normal limits were 1.02 for chil-
Brief Communication: Rust et d: CSF IgG in Children 407
Index vs. Age
Fig 2. Scattergram of CSFlserum IgG index as a function of age
for the entire reference population. Solid lines indicate 95%
confidencelimits (mean 5 two standard deviations),and
hatched lines indicate mean values.
dren less than 1.5 years old, 0.78 for older children.
Although there was a weak correlation between age
and index for the entire population ( r = 0.164, p 5
0.02, n = 195), analysis of individual age epochs failed
to isolate the age-related variation to a specific subgroup. CSF IgG/Alb ratio values correlated with IgG
index values for the entire population (r = 0.366, p 5
0.001). Children with results above our normal limits
for ratio and index are listed in the Table. The overall
rate of false positivity for CSF IgG/Alb ratio was
2.7%, and for the CSF/serum IgG index it was 2.1%.
Only U193 patients (0.5%) in whom both ratio and
index were measured showed simultaneous elevation.
None of the patients with either an elevated ratio or
an abnormal IgG index value and in whom qualitative
IgG studies were performed had oligoclonal IgG
bands. One hundred and seventy-seven children had
AGE and 95 had IEF to identify oligoclonal IgG subpopulations. Nine patients (5.1%) had false-positive
oligoclonal bands by AGE, whereas 4 patients (4.2%)
had false-positive bands by IEF. Although the mean
value for IgG/Alb ratio for patients with oligoclonal
bands was elevated (0.15 for positive AGE and 0.19
for positive IEF), no patient with false-positive oligoclonal bands had a ratio value that was greater than the
upper normal limit. The mean index values for patients
with oligoclonal bands by AGE or IEF did not differ
408 Annals of Neurology Vol 23 No 4 April 1988
from those for patients without bands. Among 175
patients with at least one quantitative (ratio or index)
and one qualitative (AGE or IEF) evaluation, none had
simultaneous abnormalities in both categories.
CSF is seldom obtained from normal children or
adults; thus reference values for CSF IgG must be
obtained from symptomatic patients without evidence
of conditions likely to provoke CSF IgG abnormality.
During early childhood, inflammatory illnesses are
common, and both blood-brain barrier efficiency and
serum IgG concentration (to which the rate of bloodbrain barrier uansudation is proportional) are subject
to marked developmental changes 11, 2). Studies of
CSF IgG that include children with acute febrile illnesses, meningism, complicated epilepsy, or CNS degenerations in the normative population 13, 6, 9-121
are thus including diagnoses associated with elevation
of CSF IgG [ S , 7, 13-15}. Studies reporting CSF IgG
but no marker for blood-brain barrier (total CSF protein or Alb) 14, 6, 10, 12, 15, 161 are difficult to
interpret. Those reporting the ratio of CSF IgG to CSF
Alb 12, 11, 141 employ a more appropriate bloodbrain barrier marker than those reporting the ratio of
CSF IgG to total CSF protein C3, 7, 8, 151 because Alb
is synthesized exclusively outside of the CNS 117, 19,
20). Only two prior studies 11, 91 report IgG data as
the CSF/serum I& index, an expression that accounts
for both blood-brain barrier and serum IgG variations
[19-2 11.
False Positives
Normal ("fatigue")
0.1 1
Myopathy (myotubular)
Head trauma
Static encephalopathy
Epilepsy (PC)
Epilepsy (PS)
Epilepsy (GTC)
"Indicatesabnormal value.
AGE = agarose gel electrophoresis;IEF = isoelectric focusing on polyacrylamide gel; PC = partial complex; PS = partial simple; GTC =
generalized tonic-clonic;
= positive; - = negative.
We suspect that the skewing of CSF IgG/Alb ratio
to the upper limits of normal prior to 4 months of age
reflects enhanced CNS access for maternally derived
serum IgG [2}, whereas the low CSF IgG/Alb ratio
during the interval 4 to 15 months of age results from
the physiological decline in serum IgG concentration.
This CSF IgG nadir has been noted in prior studies C2,
4, 5, 7, 12, 161. Higher reference values for CSF IgG
index in young children may result from selective enhancement of blood-brain barrier permeability to IgG
as compared to Alb in developing brain. Our reference
values are comparable to those found in infants by
Gerbaut and coworkers 191. As blood-brain barrier
restriction of I& and Alb uansudation is not known
to change significantly after the first few years of life
{2}, it is unclear why the upper normal limits for CSF
IgG/Alb ratio and CSF/serum IgG index in children
1.5 to 20.5 years differ from those established in a
similarly composed reference population of adults
studied by identical methods (0.25 and 0.68, respectivelv)
,, (181.
- The values determined for older children
by Gerbaut and c o u e w e s {9] differ somewhat from
ours, perhaps as a result of their inclusion of children
with febrile convulsions and CNS degenerations.
The CSF IgG/Af, ratio and C S F / 1
values in our study show significant correlation, but the
IgG/Alb ratio has a slightly higher rate of false positivity (2.7%) than does the IgG index (2.1%); only 1
patient showed false positivity for both (0.5%). Borderline values must be interpreted with caution, given
interassay and intraassay variability of up to 10% in
absolute concentration determinations by our methods, especially for specimens with very high or very
low CSF IgG concentration 1181. AGE and IEF show
higher rates of false positivity (5.1 and 4.2%, respectively) than do quantitative values. Comparable results
were found in our reference adult patients {18}. Determination of both qualitative and quantitative IgG
should be performed for each patient because simultaneous false positivity was not observed in our reference population and may be extremely rare in otherwise normal children. An isolated quantitative or
qualitative IgG abnormality should always be interpreted with caution.
Dr Rust was
in part by NINCDSgrant NS07027-11,
The excellent technical contributions of Mrs Geneva Banks and the
superb secretarial assistance of Mrs Patti Nacci are gratefully acknowledged,
Presented in part at the 15th Annual Meeting of the Child Neurology Society, Boston, Massachusetts, October 1986.
Brief Communication: Rust et
CSF IgG in Children
1. fig-Olofsson 0, Link H, Wigem A. Concentrations of CSF
proteins as a measure of blood brain barrier function and synthesis of IgG within the CNS in “normal”subjects from the age
of 6 months to 30 years. Acta Paediatr Scand 1981;70:167-170
2. Staa A, Felgenhauer K. Development of the blood-CSF barrier. Dev Med Child Neurol 1983;25:152-161
3. Harms D. Comparative quantitation of immunoglobulin G
(IgG) in cerebrospinalfluid and serum of children. Eur Neurol
4. Krause HD, Wisser H. Normalbereich des Gesamteiweisses
und der Eiweissfraktionen des Liquor cerebrospinalis bei Kindem. 2 Klin Chem Klin Biochem 1975;13:137-142
5 . Liappis N, J&el A. Normalbereich der mittels radialer I m u n diffusion bestimmren Albumjn- und IgG-Konzentrarionim Liquor Cerebrospinalis von Kindern. Klin Paediatr 1976188:
6. Mietens C, Quarcoo H. Immunoglobulin-Konzenmtionim Liquor bei Kindem Untersuchungen zur Altersabhhgigkeit und
zu Veriinderungen bei Entziindlichen Erkrankungen des ZNS.
Klin Paediatr 1977;189:151-154
7. Nellhaus G. Cerebrospinal fluid immunoglobulin G in childhood. Arch Neurol 1971;24:441-448
8. Siemes H, Siegen M, Hanefeld F. Occurrence of ohgoclonal
gammaglobulin in the CSF of children with prolonged and
chronic CNS-infections. Acta Paediatr Scand 1981;7091-99
9. Gerbaut L, Ponsot G, DuIac 0,et al. Profil prot6ique du liquide
ckphalo-rachidien chez Penfant. Determination des limites de
rkfkrence. Arch Fr Pediatr 1981;38:3-9
10. Gill DG, Brody M: Cerebrospinal fluid immunoglobulins in
children. Arch Dis Child 1979;54:961-967
11. Kariaky D. Die quantitative Bestimmung einzelner Liquorproteine in der Diagnostik entziindlicher ZNS-Erkrankungen im
Kindesalter. Eur J Pediatr 1975;121:51-57
12. MaurerJ, Rieder HP. Totalprotein und elekuophoretischeProteinfraktionen des Liquors im Kindesalter. Schweiz Med
Wochenschr 1978;108:1854- 1860
13. Hrazdirovii V, Hrazdira CL,Poltikovi M. Imunoglobuliny V
mozkom3nim moku deti upicich epilepsii. Csk Neurol Neurochk 1982;45:224-228
14. PeterslungNA, Pedersen B. Liquor:serum quotients of IgG and
albumin in patients with meningism, meningitis and multiple
sclerosis. Acta Neurol Scand I982;6625-33
15. Cesnak D, Krirlk M, Pavkovtekovi 0, et al. Diagnostickq *znam hodn6t I& V mozgovomiechovom moku pri ochoreniach
CNS u deti. Car, U k cesk 1980;119:796-799
16. diCostanzo J, Rornette J, Levy G, et al. Dosage immunonkphkl&n&rique des immunoglobulines G, A, M, de I’alpha2-macroglobuline et de la c6ruloplasmine: concentrations dans
le liquide c6phalorachidien de I’enfant et de I’adulte. Ann Biol
Clin (Paris) 1980;38213-235
17. Tourtellotte WW. Cerebrospinalfluid immunoglobulins and the
central nervous system as an immunologicalorgan particularly in
multiple sclerosis and subacute sclerosing panencephalitis. In:
Rowland LP (ed). ARNMD immunological disorders of the
nervous system, Vol 49. Baltimore: Williams & Wilkins,
18. Hershey LA, Trotter J L The use and abuse of the cerebrospinal
fluid IgG profile in the adult: a practical evaluation. Ann Neurol
19. Delpech B, Lichtblau E. Etude quantitative des immunoglobulines G et de I’albumine du liquide c6phalorachidien. Clin
Chim Acta 1972;37:15-23
20. Tibbling G, Link H, Ohman S. Principles of albumin and IgG
analyses in neurological disorders. I. Establishment of reference
values. Scand J Clin Lab Invest 1977;37:385-390
21. Trotter JL, Rust RS. Human cerebrospinal fluid (CSF) immunology. Cerebrospinal Fluid (in press)
Congenital Odontoid
A&ia and Posterior
Ckculation Stroke
in Childhood
Peter C. Phillips, MD,”? Kevin J. Lorentsen, BA,X
Lowry C. Shropshire, MD,* and Hyo S. Ahn, MD’4
Head trauma and vigorous physical activity were followed by delayed-onset posterior circulation stroke in a
5-year-old boy with odontoid aplasia and C1-C2 subluxation. Angiography demonstrated vertebral artery narrowing and occlusion of the left anterior-inferiorcerebellar and posterior cerebral arteries. Odontoid aplasia
and other atlantoaxial dislocations are treatable causes
of posterior circulation insufficiency in childhood; these
defects may be overlooked without appropriate radiographic study of the cervical spine using flexionextension views.
Phillips PC, Lorentsen KJ, Shropshire LC,
Ahn HS. Congenital odontoid aplasia and
posterior circulation stroke in childhood.
Ann Neurol 1988;23:410-413
Congenital absence of the odontoid process may lead
to serious neurological complications in adults and
children [I, 21. This infrequently recognized and potentially treatable cause of posterior circulation insufficiency in childhood may be overlooked without
radiological evaluation of cervical spine anatomy and
stability. We report a case of delayed-onset posterior
circulation (PC) stroke in a 5-year-old boy with congenital odontoid aplasia and atlantoaxial instability.
Case Report
A previously well 5-year-old right-handed boy was admitted
with ataxia and nystagmus of 7 days’ duration. Three months
earlier, while sledding, he sustained moderate head trauma
and cervical hyperextension without apparent neurological
injury. Six weeks before admission, he had severe headaches
that lasted 3 days. Three weeks later he developed acute
pallor, diaphoresis, and left arm weakness that lasted several
hours before complete resolution. Ten days before admission, after somersaulting in bed, he became ataxic, was
nauseated, and vomited. These symptoms resolved completely within 3 days.
Immediately after a vigorous bout of wrestling, he devel-
From the Departments of *Neurology, ?Pediatrics,and $Radiology,
The Johns Hopkins Hospital, Baltimore, MD.
Received Aug 4, 1987, and in revised form Sep 29 and Oct 21.
Accepted for publication Oct 23, 1987.
Address correspondence to Dr Phillips, Department of Neurology,
Johns Hopkins Hospital, 600 N Wore Street, Baltimore, MD
410 Copyright 0 1988 by the American Neurological Association
Без категории
Размер файла
497 Кб
igg, references, childhood, values, establishments, fluid, cerebrospinal
Пожаловаться на содержимое документа