вход по аккаунту


Cerebrospinal fluid vasopressin and increased intracranial pressure.

код для вставкиСкачать
Cerebrospinal Fluid Vasopressin
and Increased Intracranial Pressure
Per Soelberg Sorensen, M D , Flemming Gjerris, MD, and Mogens Hammer, MD
Cerebrospinal fluid and plasma vasopressin were measured in patients with cerebral disorders associated with varying
levels of elevated intracranial pressure. T h e mean cerebrospinal fluid vasopressin concentration was significantly
increased in patients with pseudotumor cerebri (2.0 k 0.2 {SEMI pg/ml), intracranial tumor (2.3 k 0.4 pg/ml), and
intracranial hemorrhage (1.9 k 0.3 pg/ml) compared with control patients (1.2 2 0.1 pg/ml). A significant relationship
was found between intracranial pressure and the cerebrospinal fluid vasopressin concentration within all groups of
patients and in the whole sample as well (r = 0.79; p < 0.001). I n the groups of patients with intracranial tumor,
hydrocephalus, and intracranial hemorrhage, some individuals showed plasma vasopressin concentrations inappropriate to the corresponding plasma osmolality, but no relationship was found between intracranial pressure and plasma
vasopressin concentration. I t is suggested that increased intracranial pressure is a stimulus to centrally released vasopressin. The clinical importance of increased cerebrospinal fluid vasopressin concentrations is still not known.
SGrensen PS, Gjerris F, Hammer M: Cerebrospinal fluid vasopressin and increased intracranial pressure.
Ann Neurol 15:435-440, 1084
There is increasing evidence to suggest that vasopressin
(AVP) in plasma and in cerebrospinal fluid (CSF) has
separate origins [ l l , IS, 171. In normal subjects the
CSF A V P concentration is usually lower than the corresponding plasma AVP value, and the blood-CSF barrier seems to be impermeable to the peptide in both
directions {lo, 14, 15, 261. T h e control mechanisms
and function of A V P in CSF are still largely unknown,
but vasopressinergic nerve endings have been demonstrated in various parts of the brain [23, 271, and AVP
neurosecretory pathways have been traced to the choroid plexus and the ependymal surface of the ventricles, indicating a direct release of AVP into the cerebral ventricular system [2, 31.
Recently, increased concentrations of AVP in CSF
o r plasma have been reported in a number of conditions in which elevated intracranial pressure (ICP) is a
predominant clinical feature: subarachnoid hemorrhage [l6}, bacterial meningitis 151, stroke [12), and
pseudotumor cerebri [241.
We have studied CSF and plasma AVP concentrations in cerebral disorders associated with varying,
measured levels of elevated ICP and have attempted to
correlate the measurements.
enlarged blind spots, and normal or diminished ventricular
size evident on computed tomography (CT). Their ages
ranged from 13 to 53 years (median, 37 years). Twelve of the
patients were examined during the acute phase of the disease;
4 were beginning remission. None of the patients received
any drugs during the 72 hours before the examination. CSF
samples were collected by lumbar puncture using local anesthesia, with the patients in the lateral recumbent position.
Blood samples were taken by venipuncture just before the
lumbar puncture. In 10 patients ICP was monitored for 4 to
24 hours by an epidural transducer (Philips, Eindhoven, Holland) placed through a precoronal burr hole 173, and the
mean pressure was calculated with the right atrium as reference zero level. In the others ICP was measured by lumbar
puncture after the patients had been fully relaxed and before
removal of CSF. AIL had low or normal CSF protein concentrations (0.20 to 0.50 g d L ; median, 0.35 g d L ; normal
range, 0.30 to 0.85 gm/L).
Intrucruniul Tumor
Nine patients (6 males and 3 females) had intracranial tumor:
The group with pseudotumor cerebri comprised 16 patients,
4 males and 12 females. All had raised ICP, papilledema,
5 acoustic neuromas, 1 cerebellar astrocytoma, 1 cerebellar
hemangioblastoma, 1 fourth ventricle papilloma of the choroid plexus, and 1 pinealoma. Five patients with tumors had
complicating obstructive hydrocephalus. The ages ranged
from 7 to 73 years, with a median age of 45 years. ICP was
measured in anesthetized patients via a cannula introduced
into the lateral ventricle during a shunt operation before surgical removal of the tumor. Ventricular CSF samples were
obtained through the cannula, and in all but 2 patients venous
blood samples were taken simultaneously.
From the University Clinic of Neurosurgery and Medicine P,
Rigshospitalet, DK-2 100 Copenhagen, Denmark.
Address reprint requests to Dr S@rensen, University Clinic of
Neurosurgery, Rigshospitalet, DK-2 100 Copenhagen, Denmark.
Materials and Methods
Pseudotumor Cerebri
Received May 12, 1983, and in revised form Aug 3 and Sept 16.
Accepted for publication Sept 17, 1983.
Analytical Methods
Eighteen hydrocephalic patients were studied. Fourteen, 10
rnen and 4 women aged 39 to 67 years (median, 59 years),
had normal-pressure hydrocephalus with clinical symptoms
of dementia, gait disturbances, andor urinary incontinence.
All had ventricular enlargement evident on CT, normal ICP,
and decreased conductance to CSF outflow measured by a
lumbar-ventricular perfusion study [ l}. Two women, 18 and
62 years old, had communicating high-pressure hydrocephalus as a complication of subarachnoid hemorrhage. Two children, aged 2 and 4 years, had congenital hydrocephalus and
obstruction of a previously placed ventriculoatrial shunt. In
all adult patients the intraventricular ICP was continuously
recorded during a 24-hour period by a Statham P 50 transducer. Lumbar and ventricular CSF samples were taken simultaneously before the lumbar-ventricular perfusion study,
and blood samples were drawn just before the lumbar puncture. During operative revision of the obstructed shunt in the
2 children, ICP was measured and ventricular CSF collected.
No blood samples were obtained in these 2 patients.
CSF and blood for hormone analyses were sampled in chilled
tubes and placed on ice. The blood samples were taken in
plastic tubes containing 8 mg sodium ethylenediaminetetraacetate and immediately separated at 4°C. Plasma and CSF
were stored at - 20°C until AVP analysis. AVP concentrations in CSF and plasma were measured by radioimmunoassay after extraction with acetone and petroleum ether, and all
values were corrected for the actual loss during the extraction
procedure. Each sample was extracted in duplicate, and each
sample of extracted plasma and CSF was assayed in duplicate.
The detection limit of the analysis was 0.5 pdml, and the
interassay coefficients of variation were 10 to 15%. The intraassay coefficient of variation depended on the concentration and ranged from 5 to 10%. Cross-reactions with oxytocin and vasotocin were less than 1: lo5 [8].
CSF and plasma osmolality were measured by freezingpoint depression (Knauer automated digital osmometer). The
coefficient of variation was less than 1%.
Tests for differences in hormone and osmolality levels
were made by one- or two-way analysis of variance, with a
significance level of p < 0.05.
Intracranial Hemorrhage
Nine patients, 2 men and 7 women aged 28 to 66 years
(median, 52 years), had intracranial hemorrhage: 6 had
a recent subarachnoid hemorrhage and 3 intracerebral
hematoma. In 7 patients the intraventricular pressure was
monitored continuously for at least 24 hours. CSF samples
were obtained from the ventricular system, and blood samples were taken simultaneously. In 2, the last patients, ICP
was measured and CSF samples obtained by lumbar puncture.
Control Patients
The control group included 15 patients (5 men and 10
women aged 2 1 to 6 9 years, with a median age of 39 years).
The diagnoses were headache or common migraine (8 patients), neurosis (6 patients), and paresthesia (1 patient). No
patients had signs indicating central nervous system lesions,
and no endocrine disorders were suspected. ICP was
measured by lumbar puncture; CSF and blood samples were
taken in connection with the lumbar puncture as previously
Intracranial Pressure
The mean ICP was significantly increased in the groups
of patients with pseudotumor cerebri, intracranial
tumor, and intracranial hemorrhage compared with
controls (Table). An ICP of 18 mm Hg, equivalent to
250 mm of water, was chosen arbitrarily as the iower
limit of increased ICP. Eleven of 16 patients with
pseudotumor and 6 of 9 patients with intracranial
tumor had significantly increased ICP. Fourteen patients with hydrocephalus had normal pressure, and 4
had high pressure. In the group with intracranial
hemorrhage, 6 of 9 patients had ICP equal to or greater
than 18 mm Hg. All control patients had normal ICP.
Possibly because of difficulties in obtaining full relaxation in some patients, the ICP measured by lumbar
puncture was sometimes slightly higher than the mean
Cerebrospinal Fluid Vasopressin Concentration and Osmolality, Plasma Vasopressin Concentration and Osmolality,
and Intracranial Pressure in 67 Patients"
CSF Osmolality
Plasma AVP
Plasma Osmolality
(mm Hg)
Pseudotumor cerebri (n = 16)
Intracranial tumor (n = 91b
Hydrocephalus (n = 18)'
Intracranial hemorrhage (n = 9)
Controls (n = 15)
2.0 & 0.2d
2.3 r 0.4'
1.5 r 0.2'
1.9 ? 0.3'
1.2 r 0.1
285 ? 2
285 +- 2
280 +- 2
274 i 10
281 ? 2
285 +- 2
287 ir 2
283 IT 2
280 ir 8
284 ? 2
"All values are expressed as means i SEMs.
hFor plasma AVP and osmolality, n = 7.
'For plasma AVP and osmolaliry, n = 16.
Statistical significance (difference from control group): "'p < 0.01; ' p < 0.05.
cerebrospinal fluid; A V P
vasopressin; ICP
intracranial pressure.
436 Annals of Neurology Vol 15 No 5 May 1984
ir 2d
t 5d
+- I d
N: 9
N =18
a A
N = 9
0 0 00
Fig 1. Cerebrospinaljuid vasopressin concentrations in patients
with pseudotumor cerebri, intracranial tumor, hydrocephalus,
and intracranial hemorrhage, and controls. (Open circles = patients with intracranialpressureless than 18 mm Hg; filled circles = patients with intracranial pressure equal to or greater
than 18 mm Hg.)
ICP calculated from recording of the epidural or intraventricular pressure for several hours.
Cerebrospinal Fluid Vasopressin
The mean CSF AVP concentrations were significantly
higher in patients with pseudotumor, intracranial
tumor, and intracranial hemorrhage (see the Table). In
all patients the highest CSF AVP concentrations were
associated with ICPs of at least 18 mm Hg (Fig 1). In
28 patients with such ICPs, the mean CSF AVP concentration was significantly higher (2.4 & 0.2 [SEM}
pg/ml) than in 39 patients with ICPs below 18 mm Hg
(1.2 k 0.1 pglml) (p < 0.001). Figure 2 shows a highly
significant correlation between ICP and CSF AVP concentration ( r = 0.79; p < 0.001), and this correlation
was found in all patient groups except in the control
Two hydrocephalic patients with high ICP were
reexamined after a shunt operation that normalized the
ICP. The respective ICP and CSF AVP values before
and after shunting were 35 mm Hg, 4.7 pg/ml and 8
mm Hg, 1.3 pg/ml in the first patient, and 25 mm Hg,
2.0 pglml and 14 mm Hg, 1.3 pglml in the second.
One patient developed an intracerebral hematoma as
a complication of intraventricular pressure monitoring.
Corresponding ICP and CSF AVP values were
Fig 2. Relationship between intracranial pressure (ICPI and cerebrospinaljluid vasopressin concentration in 67 patients (r =
0.79; p < 0.001):16 patients with pseudotumor cerebri
(r = 0.76; p < 0.001), 9 patients with intracranial tumor
(r = 0.80;p < 0.05), 18patients with hydrocephalus (r =
0.84; p < 0.001), 9 patients with intracranial hemorrhage
(r = 0.69; p < 0.0>), and 15 controls (r = 0.35; p > 0.1).
Open symbols indicate epidural or intraventricular ICP measurement;&llea!symbols indicate ICP measured by lumbar
measured before (7 mm Hg, 1.0 pglml) and immediately afterward (25 mm Hg, 1.6 pg/ml), during
neurosurgical removal of the hematoma (25 mm Hg,
2.7 pglml), and the day after the operation (15 mm Hg,
0.7 pglml).
In 14 patients with normal-pressure hydrocephalus,
samples of ventricular and lumbar CSF were taken simultaneously, and no significant difference in AVP
concentration was found between ventricular CSF (1.6
% 0.4 pglml) and lumbar CSF (1.4 f 0.5 pglml).
Plasma Vasopressin and Osmolality
The mean plasma AVP concentration was significantly
increased in patients with intracranial hemorrhage (p <
0.05) compared with the control patients. All control
patients had plasma AVP concentrations appropriate to
the plasma osmolality (Fig 3). This normal relationship
was also found in the patients with pseudotumor. Of
the other patients, some showed plasma AVP concentrations inappropriately high or low in relation to the
corresponding plasma osmolality (see Fig 3). The inappropriate plasma AVP concentrations were found in
patients with raised as well as normal ICP, and no
significant relationship was found between ICP and
plasma AVP @ > 0.05). CSF AVP and plasma AVP
were significantly correlated in patients with pseudotumor (p < 0.05), whereas this relationship was not
found in the other groups (Fig 4).
Sorensen et aI: CSF AVP and Increased ICP 437
hi : 16
N i 16
N =16
. ..
Fig 4. Relationship between plasma and cerebrospinaljuid uaso-pressin Concentrations in 16 patients with pseudotumor cerebri
(r = 0.61; p < 0.051, 7 patients With intracranial tumor (r =
-0.12: p > 0,1), Ibpatients with hydrocephalus (r =
-0.17; p > 0.1), 9 patients with intracranial hemorrhage
(r = 0 . 1 8 ; >
~ O . l ) , und I5 controls Ir = 0 . 1 0 ; ~
> 0.1).
Fig 3. Relationship between plasma osmolality and plasma vasopressin concentration in 16 patients with pseudotumor cerebri, 7
patients with intracranial tumor, 16 patients with hydrocephalus, 9 paiients with intracranial hemorrhage, and 15 controls.
The total range in normal subjects is indicated by the area between solid fines {9},
Mean plasma osmolality showed no significant variation between the patient groups, but plasma hypoosmolality or hyperosmolality was observed in some
patients with intracranial hemorrhage and hydrocephalus.
The present study has demonstrated increased CSF
AVP levels in patients with several cerebral diseases
associated with increased ICP. A significant relationship between ICP and CSF AVP was found within all
patient groups, and in the whole sample as well. In the
control group ICP and CSF AVP were randomly distributed within the normal limits. In patients with
pseudotumor and in controls, AVP was measured in
lumbar CSF, whereas ventricular CSF was used in the
majority of patients with intracranial hemorrhage or
tumor because of the hazards of lumbar puncture.
438 Annals of Neurology Vol 15 No 5 May 1984
Comparison of AVP concentrations in CSF does not
seem to be affected by the level within the neuroaxis at
which the CSF is obtained, however. In our patients
with hydrocephalus we found no difference in AVI)
concentrations between lumbar and ventricular CSI;
specimens obtained simultaneously. This result accords
with the findings of Luerssen and Robertson 1141 that
the concentration of AVP from the lumbar cistern
closely approximates that from more rostral parts of
the neuroaxis. Serial assays on three 5 ml fractions of
CSF tapped from the lumbar cistern in patients prior to
myelography indicated no rostrocaudal gradient; the
mean AVP concentrations ( +- SEM) were: 0 to 5 ml,
1.3 pg/ml f 0.1; 5 to 10 ml, 1.4 pg/ml -+ 0.1; and 10
to 15 ml, 1.4 pg/ml t 0.1 {lo}.
A number of previous observations are consistent
with a relationship between elevated ICP and increased
CSF AVP. Mather and colleagues 1161 reported increased CSF AVP in 5 patients with subarachnoid
hemorrhage and severely disturbed consciousness. No
information was given concerning the ICP in these 5
patients, but increased ICP is a well-known concomitant of severe subarachnoid hemorrhage {25}. Garcia
and colleagues [5} found significantly increased CSF
AVP in children with bacterial meningitis; posttreatment CSF AVP concentrations were lower than thost.
observed on admission. Reid and Morton {22) have
corroborated our previous finding of elevated CSE;
AVP in benign intracranial hypertension (pseudotumor cerebri) [241. The overall findings indicate
that increased ICP is a stimulus to centrally released
Plasma AVP was also higher in our groups of patients with intracranial tumor and intracranial hemorrhage, but a relationship between plasma AVP concentrations and ICP could not be demonstrated. However,
in all groups of patients except the controls, some patients showed elevated plasma AVP concentrations inappropriate to the plasma osmolality (91. Others have
demonstrated high plasma AVP levels in patients with
intracranial disorders. Wise 128) reported inappropriate secretion of antidiuretic hormone caused by obstruction of ventriculoatrial shunts in 2 patients, and
Kaplan and Feigin [13) found increased plasma AVP
concentrations in children with bacterial meningitis. In
patients with subarachnoid hemorrhage, high plasma
AVP levels were associated with both hyponatremiahypo-osmolality and normonatremia { 161.Significantly
increased plasma AVP levels without accompanying
hyponatremia have been reported in patients with
stroke [12) and malignant hypertension 119). With the
exception of patients with pseudotumor, we found no
correlation between CSF AVP and plasma AVP. The
blood-brain barrier appears to be bidirectionally impermeable to AVP, and plasma and CSF AVP appear
to be controlled by different mechanisms.
The findings of a number of animal experiments support the conclusion that elevation of ICP stimulates
AVP release. In rhesus monkeys elevation of ICP produced by subdural balloons resulted in an increase in
urinary AVP excretion proportional to the level of intracranial hypertension [6]. Elevation of ICP in cats
induced by injection of homologous blood into the
subarachnoid space or by cerebral balloon compression
evoked a marked (2- to 100-fold) rise in plasma AVP
concentration {211. These results contrast with our
finding of a normal plasma AVP in most patients with
increased ICP.
It is not known whether increased CSF AVP concentrations have clinical importance. Raichle and
Grubb [ZO) demonstrated that the level of intraventricular AVP influences brain water permeability in
rhesus monkeys, and suggested that disturbances in the
central vasopressinergic system relate to the development of brain edema. Intraventricular administration of
AVP increased the brain water content in rats without
inducing great changes in electrolyte concentrations in
the accumulated fluid (4). We have injected AVP concentrations of 0.1 to 3 ng/ml into the lateral ventricles
of 4 rabbits and produced elevated ICP (unpublished
data). Noto and colleagues [18), however, observed a
decrease in the ICP in rabbits after intraventricular
AVP injection. The relationship of CSF AVP to brain
edema is at best speculative.
Supported by grants from the Danish Medical Research Council.
1. Bargesen SE, Gjerris F, S0rensen SC: The resistance to cerebrospinal fluid absorption in humans. Acta Neurol Scand 5 J:88-96,
2. Brownfield MS, Kozlowski GP: The hypothalamo-choroidal
tract. I. Immunohistochemical demonstration of neurophysin
pathways to telencephalic choroid plexuses and cerebrospinal
fluid. Cell Tissue Res 178:lll-127, 1977
3. Buijs RM, Swaab DF, DogteromJ, Van Leeuwen FW:Intra- and
extrahypothalamic vasopressin and oxytocin pathways in the rat.
Cell Tissue Res 186:423-433, 1978
4. Doczi T, Szerdahelyi P, Gulya K, KISSJ: Brain water accumulation after the central administration of vasopressin. Neurosurgery 11:402-407, 1982
5. Garcia H, Kaplan SL, Feigin RD: Cerebrospinal fluid concentration of arginine vasopressin in children with bacterial meningitis.
J Pediatr 98:67-70, 1981
6. Gaulin L, Skowsky WR, Goodman SJ: Release of antidiuretic
hormone during mass-induced elevation of intracranial pressure.
J Neurosurg 46:627-637, 1977
7. Gjerris F, Borgesen SE, Sorensen SC: Clinical evaluation of a
new epidural pressure transducer. In Shulrnan K, Becker DP,
Hochwald GM, Brock M (eds): Intracranial Pressure, Vol 4.
Berlin, Heidelberg, New York, Springer-Verlag, 1979, pp 386-
8. Hammer M: Radioimmunoassay of 8-arginine-vasopressin(antidiuretic hormone) in human plasma. Scand J Clin Lab Invest
38:707-716, 1978
9. Hammer M, Ladefoged J, Blgaard K: Relationship between
plasma osmolaliry and plasma vasopressin in human subjects.
Am J Physiol 238:E313-E317, 1979
10. Hammer M, Sprrensen PS, Gjerris F, Larsen K: Vasopressin in
the cerebrospinal fluid of patients with normal pressure hydrocephalus and benign intracranial hypertension. Acta Endocrinol
100:211-215, 1982
11. Jenkins JS, Mather HM: Ang V: Vasopressin in human cerebrospinal fluid. J Clin Endocrinol Metab 50:364-367, 1980
12. Joynt RJ, Feibel JH, Sladek CM: Antidiuretic hormone levels in
stroke patients. Ann Neurol 9:182-184, 1981
13. Kaplan SL, Feigin RD: The syndrome of inappropriate secretion
of antidiuretic hormone in children with bacterial meningitis. J
Pediatr 92:758-761, 1978
14. Luerssen TG, Robertson G L Cerebrospinal fluid vasopressin
and vasotocin in health and disease. In Wood J H (ed):
Neurobiology of Cerebrospinal Fluid, Vol 1. New York,
Plenum, 1980, pp 613-623
15. Luerssen TG, Shelton RL, Robertson GL: Evidence for separate
origin of plasma and cerebrospinal fluid vasopressin. Clin Res
25:14A, 1977
6. Mather HM, Ang V, Jenkins JS: Vasopressin in plasma and CSF
of patients with subarachnoid hemorrhage. J Neurol Neurosurg
Psychiatry 44:216-219, 1981
7. Mens WBJ, Bouman HJ, Bakker EAD, Van Wimersma
Greidanus TB: Differential effects of various stimuli on AVP
levels in blood and cerebrospinal fluid. Eur J PharmacoA68:8992, 1980
8. Noto T, NakajimaT, Saji Y,Nagawa Y .Effect ofvasopressin on
intracranial pressure of rabbit. Endocrinol Jpn 25:591-596,
19. Padfield PL, Brown JJ, Lever AF, et al: Does vasopressin play a
role in the pathogenesis of hypertension? Clin Sci {Suppl]
61:14 IS- 1 4 3 ~198
, 1
20. Raichle ME, Grubb F
XJr: Regulation of brain water permeabil-
Sarensen et al: CSF AVP and Increased ICP
2 1.
ity by centrally-released vasopressin. Brain Res 143:191-194,
25. Voldby B, Enevoldsen EM: Inrracranial pressure changes follorv-
Rap ZM, Chwalbinska-Moneta J: Vasopressin concentration in
blood during acute short-term intracranial hypertension in cats.
Adv Neurol 20:381-388, 1978
Reid AC, Morton JJ: Arginine vasopressin levels in cerebrospinal fluid in neurological disease.J Neurol Sci 54:295-301, 1982
Rossor MN, Iversen LI., Hawthorn J, et al: Extrahypothalamic
vasopressin in human brain. Brain Res 214:349-355, 1981
Sorensen PS, Hammer M, Gjerris F: Cerebrospinal fluid vasopressin in benign intracranial hypertension. Neurology (NY)
32:1255-1259, 1982
ing aneurysm rupture. J Neurosurg 56:186-196, 1982
26. Vorherr H , Bradbury MWB, Hoghoughi M, Kleeman CK:
Antidiuretic hormone in cerebrospinal fluid during endogenous
and exogenous changes in its blood level. Endocrinology
83:246-250, 1968
27. Weindl A, Sofroniew MV: Immunohistochemical localization of
hypothalamic peptide hormones in the neural target areas. In
Wuttke A, Weindl A, Voight KH, Dries RR (eds): Brain and
Pituitary Peptides. Basel, Karger, 1980, pp 97-109
28. Wise BL: Inappropriate secretion of ADH caused by obstruction of venrriculoatrial shunts. J Neurosurg 28:429-432, 1968
440 Annals of Neurology Vol 15 No 5 May 1984
Без категории
Размер файла
522 Кб
vasopressin, increase, pressure, intracranial, fluid, cerebrospinal
Пожаловаться на содержимое документа