close

Вход

Забыли?

вход по аккаунту

?

Central pontine myelinolysis following rapid correction of hyponatremia.

код для вставкиСкачать
ORIGINAL ARTICLES
Central Pontine Myelinolysis Following
Rapid Correction of Hyponatremia
Robert Laureno. MD
Central pontine and extrapontine myelinolysis was experimentally produced in dogs by the rapid correction of severe,
sustained, vasopressin-induced hyponatremia. Hyponatremia alone or slowly corrected hyponatremia did not produce
the disease. Affected dogs showed rigid quadriparesis. The central pons, lateral aspects of the thalamus and adjacent
internal capsules, deep layers of cerebral cortex and subjacent white matter, cerebellum, and other regions were
symmetrically involved. Myelin and oligodendroglia were affected out of proportion to axons and neurons. Thus, the
clinical features, the distribution of the lesions, and their histological features closely resemble the human disease.
These experiments document an electrolyte manipulation that can cause permanent neuropathological lesions. Taken
with the available clinical data on human patients, the experimental results indicate that human myelinolysis may be
due to a rapid increase in serum sodium from previously low levels, and that rapid normalization of severe, sustained
hyponatremia should therefore be avoided.
Laureno R: Central pontine myelinolysis following rapid correction of hyponatremia.
Ann Neurol 13:232-242, 1983
Central pontine myelinolysis (CPM) is a symmetrical
focus of myelin destruction at the center of the basis
pontis. Adams, Victor, and Mancall f31 carefully selected the name for this entity to indicate its stereotyped location and unique pathological picture. Nerve
cells and axis cylinders are remarkably spared within
the demyelinated zone, except at its center. Loss of
oligodendroglia, sparing of blood vessels, and absence
of inflammation are other notable features of the lesion. Large myelinolytic lesions result in subacute
quadriplegia and pseudobulbar palsy, but small lesions
may cause no symptoms. Occasionally CPM extends
into the pontine tegmentum, causing additional
neurological signs. From the symmetrical distribution
and specific localization of this pontine lesion, Adams
and his associates [3}inferred a metabolic origin. Since
all 4 of their patients were alcoholic or malnourished,
they considered a nutritional deficiency the most likely
cause.
Subsequent to its initial description, CPM has been
recognized worldwide. Reported cases now exceed
150, including children as young as 3 years. In one
series of 3,548consecutively autopsied adults, 9 cases
of CPM were found (0.25p incidence) [45}.Study of
this cumulative material has brought a few fundamental
advances in our understanding of the disease.
The recognition that 1Oc4 of reported cases show
symmetrically placed extrapontine lesions expanded
our concept of the disorder. Histologically identical to
the pontine lesions, extrapontine myelinolysis occurs in
the thalamus and adjacent internal capsule, lower levels
of cerebral cortex and subjacent white matter, cerebellar white matter, and other regions. In such instances,
altered levels of consciousness have been reported.
The features of the pontine-extrapontine form of the
disease have recently been discussed in detail 1471.
A second advance has been the increasingly strong
association of CPM with derangements of serum
sodium, usually hyponatremia [1,9, 11, 13, 22, 27, 37,
40,471.Wright, Laureno, and Victor f47fcorrelated
the time of occurrence of the sodium derangements
with the histological age of the pontine lesions. Old
lesions (prominent fibrillary gliosis) occurred in patients whose hyponatremia had antedated death by
months; relatively recent lesions, lacking fibrillary
gliosis, were found in patients whose sodium derangements had preceded death by much shorter intervals.
In addition, review of the most severe forms of the
disease-those with pontine and extrapontine involvement-also supported the importance of this association; most severe disorders of serum sodium were
characteristic of such cases 147). Some authors have
suggested that vigorous correction of hyponatremia
may be more damaging than the hyponatremia itself.
From the Departments of Medicine and Neurology, The Washington Hospital Center, and the Department of Neurology, The
George Washington University School of Medicine, Washington,
D C.
Received Feb 17, 1982, and in revised form June 14. Accepted for
publication June 21, 1982.
232
Department of Neurology,
Address reprint requests to Dr
The Washington Hospital Center, 110 Irving St, N W , Washington,
D C 20010.
For example, Tomlinson e t a1 [42) described 2 severely
hyponatremic patients in whom clinical signs of pontine and extrapontine myelinolysis (autopsy proved)
appeared only after vigorous administration of 3%
saline. Although Tomlinson's g r o u p were the first t o
emphasize this sequence of events, they noted that
others [13, 18, 37) had made similar observations.
Wright e t a1 described t h e same sequence in their patient with pontine and extrapontine involvement; they
also noted that vigorous corrective efforts for severe
hyponatremia had b e e n documented in many of the
cases with extrapontine involvement recorded in the
literature C47).
The foregoing observations led t o experiments in
which rapid correction of hyponatremia reproduced
pontine and extrapontine myelinolysis in dogs r28, 297.
Corroboration of these results came with similar experiments in which extrapontine without pontine lesions
were produced in rats [24, 261. The following report
details and analyzes the findings in experimental pontine and extrapontine myelinolysis.
Materials and Methods
Sixty mongrel dogs, weighing between 14 and 18 kg, were
given vasopressin injections and water infusions in order to
induce hyponatremia. Vasopressin tannate in oil (Pitressin)
was injected intramuscularly. Water was infused intraperitoneally; 100 to 200 mg of intravenous ketamine hydrochloride (Ketalar) was needed to sedate dogs for the water
administration. Each dog was given 50 pressor units (10 cc) of
vasopressin at the onset of an experiment. O n each of the
next two days, 50 units of vasopressin was again injected and
two liters of water infused; on subsequent days, 40 units of
vasopressin and one liter of water were administered. This
methodology varied somewhat under certain circumstances.
Occasionally the abdominal cavity could not accommodate a
full two liters of fluid. Some dogs did not develop significant
hyponatremia o n the standard regimen, in which case they
were treated with higher doses of vasopressin and water as
indicated by the serum sodium level, their clinical condition,
or both. The object was to produce severe hyponatremia
(sodium level below 120 mEq/L) for several days. (Normal
serum sodium values in the dog are 137 to 149 mEq/L 1321.)
Serum sodium levels were monitored daily. Serum potassium was monitored in many dogs. These electrolytes were
measured by a Technicon SMAC multichannel analyzer (ionselective electrode method). When the serum sodium
reached very low levels (110 mEq/L) or when the dog grew
stuporous from hyponatremia, the daily doses of vasopressin
and water were often lessened to minimize deaths from water
intoxication.
Those dogs which survived a period of sustained hyponatremia (at least four days) were managed in one of two ways.
In one group, vasopressin was terminated and absolute water
restriction was begun. In these dogs the serum sodium was
allowed to correct itself spontaneously and gradually. Autopsies were performed weeks to months after termination of
the vasopressidwater treatments. In the second group, vasopressidwater treatments were discontinued and the dogs
were infused intravenously with 3% saline (50 to 200 ml)
over a period of six hours. There was absolute water restriction until the serum sodium reached normal levels. Autopsies
were performed days to months after the saline infusions.
Dogs unable to eat and drink following normalization of
serum sodium received maintenance fluids intravenously
with frequent doses of multivitamins, including thiamine. If
there was evidence of respiratory infection, benzathine
penicillin was administered intramuscularly.
Another control group consisted of 6 dogs that were never
hyponatremic but received 300 ml of 3% saline intravenously over two to five hours. Sacrifice and autopsy were
performed weeks to months after the infusion.
Autopsies were performed either at death or at sacrifice
(with intravenous pentobarbital). Brains were fixed in 10%
formalin and were sectioned and examined grossly. Microscopic sections of thalamus and basal ganglia, multiple areas
of cerebral cortex, midbrain, multiple levels of pons, medulla,
and cerebellum were studied routinely with hematoxylin and
eosin, Bodian, lux01 fast blue, and Holzer stains.
Results
Of 60 severely hyponatremic dogs, 38 died or were
sacrificed during t h e course of water and vasopressin
administration. Most died with acute water intoxication
of one or two days' duration. Many of the dogs were
drowsy, stuporous, or comatose prior t o death; many
convulsed. None showed spasticity or rigidity. Autopsies were performed on the 6 dogs that survived at least
four days of severe hyponatremia prior t o death or
sacrifice (Table 1). N o n e showed myelinolysis. A detailed case summary of o n e of these dogs has been
reported previously 1291.
Sixteen dogs with sustained hyponatremia were
managed with corrective hypertonic saline therapy followed by absolute water restriction. T e n survived the
saline infusion by more than three days. Five of these
survivors improved clinically with normalization of the
serum sodium; n o n e showed myelinolysis at autopsy
(Table 2). With a rise in t h e serum sodium concentration, t h e o t h e r 5 dogs showed prominent neurological
deterioration, usually two days after the saline infusion
Table 1. Dogs Dying during Sustained Hyponatremia
M yelinolysis
Serum Sodium (mEq/L)
Dog
Day
NO.
-3
1
2
124
114
3
4
106
113
5
6
102
99
Day
-2
Day
106
103
105
111
107
94
101
111
107
99
97
-ld
96
Day
Ob
Pontine
Extrapontine
103
91
101
106
104
102
0
0
0
0
0
0
0
0
0
0
0
0
"Day prior to death.
bDay of death.
Laureno: Central Pontine Myelinolysis
233
Table 2. Dogs Neurologtcallj Stable or Iviprowd after Hj9ertonic Saline Therapji for Hyponatremza
Serum Sodium (mEq/L)
Dog
No.
Day
Day
Day
Day
Day
Day
Da\i
-3
-2
-1
0"
1
2
4
7
116
126
122
133
129
121
134
119
133
116
123
128
111
135
110
113
135
107
136
106
140
145
118
140
121
...
160
125
148
116
...
8
9
10
11
~
...
137
...
...
Myelinolysis
Autopsy
Day
Pontine
Extrapontine
75
36
135
29
61
0
0
0
0
0
0
0
0
0
0
~~
"Day of hypertonic saline therapy, termination o f vasopressin, and onset of water
'Day following hypertonic saline therapy
restriction
Table 3. Dogs Neurologicalb Wov.te after Hypertonic Saline Therapy for Hyponatremia"
Serum Sodium (mEq/L)
Dog
No.
Day
12
13
14
15
16
106
130
107
100
106
-3
Day
Day
Day
Day
0"
1'
2
4
Autopsy
Day
98
102
90
99
...
149
120
116
122
139
100
106
131
96
157
153
145
135
142
18
67
5
8
7
Day
-2
Day
-1
108
122
102
116
126
97
100
...
...
120
130
Myelinoly sis
Pontine
Extrapontine
+
+
+
+
+
+
0
+
+
+
"Includes only dogs that survived hypertonic saline by at least three days.
"Day of hypertonic saline therapy, termination o f vasopressin, and onset of water restriction.
'Day following hypertonic saline therapy.
(Table 3); these 5 dogs all showed myelinolytic lesions
at autopsy. The following animal is representative of
the latter group.
Dog 12
After o n e week of severe hyponatremia, this animal received
an intravenous infusion of 200 ml of 3% saline over a period
of five hours. Absolute water restriction followed. Twentyfour hours later the d o g stood unsteadily, wagging its tail.
T w o days after treatment the sodium had risen t o 139 mEq/L,
but the dog deteriorated neurologically. He was drowsy and
unable to stand. Eyelids blinked occasionally but there was no
spontaneous limb movement. T h e forelimbs withdrew from
tactile stimulation; the hindlimbs did not. T h e following day
the sodium level reached 149 mEqiL, hut the d o g remained
on his side with frequent spontaneous myoclonic jerks of all
four limbs. O n the fourth day after saline infusion (with the
sodium level at 157 mEq/L), the dog was stuporous but
blinked to threat and held all four limbs rigidly extended. O n
day 10 painful stimuli resulted in reflex extensor spasm of all
four limbs. O n day 14 improvement was evident; the clog
licked and cried spontaneously and his eyes were open, but
h e remained o n his side. O n day 18 h e was still unable to
stand. He was awake and his forelimbs movcJ only feebly,
but there were abundant, spontaneous, running movements
of the hindlimbs. Pupillary responses, eye movements, and
respirations were natural. T h e dog would cry, but he could
234 Annals of Neurology
Vol 1 3 No 3 March 198.3
not drink. He blinked to threat. He was sacrificed and autopsied.
The foregoing case is typical of these 5 dogs (see
Table 3) in that neurological deterioration became evident two or three days following infusion of saline and
institution of water restriction. With the early rise in
serum sodium, 2 dogs that had been unable to stand
during severe hyponatremia once again could stand;
this initial improvement was then followed by substantial worsening. Only 1 animal convulsed during .this
phase of clinical deterioration. Three of the 5 dogs
were stuporous or comatose. None of the 5 was able to
stand; all showed leg weakness. Four animals had intermittent extensor spasms of all four limbs, sometimes
precipitated by loud sounds or stroking of the fur.
Slow, deep, regular respirations were prominent in 2
dogs. With gradual improvement of the neurological
disorder, 2 dogs showed unequivocal ataxia. Although
serum sodium was excessively corrected in Dogs 12
and 13, the corrected sodium level never exceeded 145
mEqiL in Dogs 14, 15, and 16.
In all 5 cases the external appearance of the brain
was normal, the superior sagittal sinus was unobstructed, and the major arteries were patent. Coronal
F i g 1 . Centralpontine myelinolysis in dog and human. (A)A
section of pons from Dog 12, stained for m y e h (hxolfast blue).
(B1 For comparison, a section of human pons showing the disease.
sections of the brain appeared normal except in Dog
12, in which areas of roughened texture and discoloration were evident in the cerebral cortex and in the
lateral aspects of the thalamus. Brainstem sections in
this case showed a gray area in the central pons slightly
below the midbrain.
In 4 cases, stained sections revealed a pontine lesion
symmetrically placed on the midline (Fig 1). In all 5
dogs there were lateral thalamic lesions which involved
adjacent internal capsule (Fig 2). Cerebral cortical lesions affected the deeper layers of cortex and subjacent
white matter in patches (Fig 3). There were also more
or less symmetrical lesions of cerebellar white matter,
caudate, putamen, globus pallidus, and red nuclei. Histologically, the lesions were identical in all regions. Demyelination and loss Of oligodendrodial cells were the
most important changes. Axons were little affected except at the center of the lesions, where loss of axons
and neurons was evident. More peripherally, retained
neurons stood out in the sea of macrophages, reactive
microglia, and swollen astrocyte nuclei (Fig 4 ) that ex-
Fig 2. Symmetrical mjdiinolytic bions of the thalumuj in dog
and human. (A) Coronal .section through the thalamu.i of Dog
12, stained for myelin Iluxolfust blur). IB) For covipurison, a
of human thalamus shou,ing
the dj.sea.se,
Fig 3. A sertion of the frontal lobe of Dog 12, stairzedfir myelin
fluxolfast bluet. There are extensive patches of myelinobsis in the
cerebral cortex and .iubjacent white matter.
Laureno: Central Pontinc Myclinolysis
235
tended throughout the lesions. Purkinje cell loss was
associated with the areas of severe cerebellar white
matter involvement. In more acute examples of the
disease, in which dogs had survived shorter intervals,
there were necrotic neurons and vasculoendothelial
proliferation in the very center of the lesions. Dog 13,
whose survival was prolonged (67 days), showed
fibrillary gliosis in the damaged areas. An account of
the chronological variations in the abnormalities has
been presented previously [28]. The lesions were free
of inflammation or any abnormality of blood vessels. In
each case the hippocampus was normal, as was the corpus callosum.
The distribution of lesions was similar in all 5 dogs of
this group (Table 4 ) . Although the cerebellar lesions
Fig 4. A section through the pontine lesion from Dog 12. Arr w s point to intact neurons in a sea ofmacrophages. IH&E;
x 128.)
were maximal in the midline, they involved lateral portions of the cerebellar hemispheres as well. Dog 13 was
exceptional in having extrapontine lesions without a
pontine lesion. The pontine lesion in Dog 15 was unusual in that it extended to involve both the red nuclei
above and the upper medulla below (Fig 5 ) ; in the
other 4 dogs, the lesions of the red nuclei were separate from the pontine lesion and the medulla was not
affected.
Six of the 16 hyponatremic dogs treated with hyper-
Table 4. Distribution of Lesions in Myelinolysis after Hypertonic Saline Therapy for Hyponatremia
Dog
No.
12
13
14
15
Thalamus/
Internal Capsule
+
+
+
+
0
+
+
+
16
+
Central
Pons
+
Globus
Pallidus
Cerebral
Cortex
+
+
+
t
0
0
0
=
n o myelinolysis; . . .
236 Annals of Neurology
Vol 13 No 3
=
Neostriatum
+
+
+
+
+
+
+
myelinolysis present; 0
Cerebellum
=
+
...
+
region not available for study.
March 1983
+
+
+
+
Midbrain
Medulla
Cervical
Spinal Cord
+
0
0
0
0
+
+
+
0
+
0
0
0
0
the dogs whose autopsies showed myelinolysis. With
an increase in serum sodium, 2 dogs had clinical improvement followed by worsening. The third dog became stuporous one day after the saline infusion. These
3 dogs were found dead two or three days following
the saline infusion, too early for them to have shown
fully developed myelinolytic lesions. However, there
was a prominent degree of vasculoendothelial proliferation in the lateral thalamus. This reaction may have
represented the earliest microscopic sign of the disease.
Six severely hyponatremic dogs were treated with
absolute water restriction but without saline infusion.
The serum sodium reached normal levels in 3 of these
dogs by the time of autopsy; in the other 3 the serum
sodium rose but never surpassed 130 mEq/L. Five of
the 6 dogs remained clinically stable or improved with
this management. None of these 5 showed myelinolysis at autopsy; the case summary of a representative
animal was presented in a previous publication [29].
One of the 6 dogs worsened neurologically as the
serum sodium level increased. Despite normal serum
sodium values, he was lethargic and unable to walk for
over a week. This dog eventually recovered normal
neurological function, but autopsy revealed areas of
mild, symmetrical myelinolysis in the lateral thalamus.
There was no pontine myelinolysis. Table 5 gives the
serum sodium values for these animals during the first
week of the experiments.
Six normal dogs that were never hyponatremic were
infused intravenously with 300 ml of hypertonic saline.
In 5 the infusion elevated the serum sodium between 4
and 10 mEq/L, and in the sixth the serum sodium tose
17 mEq/L. None of these dogs showed behavioral or
postmortem abnormalities.
B
Fig 5 . Sections ofpons and medulla oblongata from Dog 1 5 . (A)
Central pontine myelinolysis. Note the prominent midline cerebellar myelinolysis. (B) Extension of the pontine lesion into the
medulla oblongata. (Both luxol fast blue.)
tonic saline failed to survive the infusion by three days.
In 1 case, the serum sodium did not rise appreciably
and the dog died in a state of persistent, severe hyponatremia; there was no myelinolysis at autopsy. Two
dogs were found dead shortly after saline therapy; no
myelinolysis was seen at autopsy. In the remaining 3
dogs, however, the clinical course was similar to that of
Discussion
The foregoing data document an expwimental neurological disorder which emerges with the correction
of sustained, vasopressin-induced hyponatremia and
which is due to symmetrical myelinolytic lesions in the
Table 5 . Hyponatremic Dogs Treated with Water Restriction
Serum Sodium (mEq/L)
Day
No.
Day
-3
Day
-2
Day
-1
Day
0”
Day
1
Day
2
4
17
18
19
20
21
22
110
113
106
106
115
107
105
111
112
115
110
102
...
113
117
106
104
102
120
118
109
100
102
107
130
116
120
109
117
114
132
119
126
113
130
118
135
121
130
112
131
114
Dog
Autopsy
Day
43
10
10
9
62
15
M yelinoly sis
Pontine
Extrapontine
0
0
0
0
0
0
0
0
0
0
+
0
“Day of onset of absoiute water restricrion and termination of vasopressin.
bDay following onset of water restriction.
Laureno: Central Pontine Myelinolysis
237
brain. The clinical features of this experimental disease,
the distribution of the lesions, and their histological
characteristics closely resemble the features of CPM in
humans. The paralysis, rigidity, and stupor produced in
these dogs are prominent features of the human disease. Also similar to the human disorder are the stereotyped localization and symmetry of the lesions in the
pons (see Fig l), lateral thalamus (see Fig 2), deep
layers of cerebral cortex, cerebellar white matter, red
nucleus, and corpus striatum. Microscopically, the experimental lesions duplicate the features of the disease
in humans: noninflammatory demyelination, relative
sparing of neurons and axis cylinders, loss of oligodendroglia, and a reaction consisting of pleomorphic microglia, macrophages, and swollen astrocyte nuclei. Deviation of the microscopic picture from the typical
human case could be attributed to the age of the lesions. An acute case showing necrotic neurons and a
very chronic case showing fibrillary gliosis are variant
features consistent with the chronology of the lesions.
Although uncommon, these features have occasionally
been described in human material.
Clinicopathological correlations are also similar to
those drawn from the human disease. The paresis and
rigidity can be attributed to interruption of corticospinal tracts in thalamocapsular and pontine lesions,
and the alteration of consciousness to the cerebrocortical, thalamic, and pontine tegmental involvement.
The cerebellar lesions account for the ataxia observed
in 2 dogs in the recovery phase. The cerebral cortical
lesions adequately explain the seizures that occurred in
1 case.
Cause of Experimental Myelinolysis
The most striking feature of this study is that with a
single exception, myelinolysis occurred only in dogs in
which the sustained hyponatremia was treated with hypertonic saline. That exception was Dog 21, in which
the serum sodium level rose rapidly with water restriction alone. No myelinolytic lesions were found in the
untreated hyponatremic dogs or in the normal dogs
that were infused with 3% saline in volumes greater
than those used to correct hyponatremia. There was n o
difference in the severity of hyponatremia (day 0) between the group of dogs treated with water restriction
and the group that developed myelinolysis following
hypertonic saline therapy. The dogs treated with water
restriction (see Table 5 ) showed a mean serum sodium
level of 109 mEq/L (range, 100 to 120 mEq/L); the
dogs with myelinolysis after hypertonic saline therapy
(see Table 3) showed a mean level of 106 mEq/L
(range, 96 to 131 mEq/L). Thus, the severity of hyponatremia does not distinguish a low-risk group from
the severely affected animals. Additionally, the case of
Dog 13 shows that the antecedent hyponatremia need
not be severe for myelinolysis to result; rather, the rate
238 Annals of Neurology Vol 1 3 No 3 March 1383
Table 6. Incidence of Myelinolysis at Autopsy
in Different Experimental Groups
Group
Group A: uncorrected
hyponatremia
Group B: hypertonic saline
infusion in normal
dogs
Group C: hyponatremia treated
with water
restriction
Group D: hyponatremia treated
with hypertonic
salinea
Groups C
and D: one-day rise in serum
sodium:
15 mEq/L
< 15 mEq/L
two-day rise in serum
sodium:
a 20 mEq/L
< 20 mEq/L
Total
No. of
Dogs
No. with
Myelinolysis
6
0
6
0
6
1 17%)
10
5 50%)
s (7157)
7
8
0
5
4 180%)
0
8
"Includes only dogs surviving saline infusion for ar least three days.
of correction of hyponatremia appears to be the critical
variable. If one combines the two treatment groups
(Table 6), it becomes evident that the rapid rise in
serum sodium either one or two days after the onset of
treatment of hyponatremia characterizes the high-risk
group.
Several other points deserve comment. Since the duration of hyponatremia was similar in all dogs in this
study, no statement can be made about the possible
importance of variations in this factor in the genesis of
the disease. Second, the hypokalemia which so often
accompanies hyponatremia does not appear to be a
factor; previous studies have shown that myelinolysis
may be induced when the serum potassium concentration is maintained at normal levels throughout the experiment [29]. Finally, it remains to be investigated
whether the occurrence of myelinolytic lesions relates
to a shift in osmolarity or whether it is more specifically
linked to changes in the sodium ion concentration. In
some situations, the sodium concentration of body
fluids relates to neurological disorders independently
of osmolarity C15, 18, 461.
Cause of Human Myelinolysis
As early as 1961, J. H. Adams [11 reported 2 cases of
CPM associated with severe hyponatremia. Numerous
examples of this association with serum sodium derangements followed [9, 22, 27, 37, 401; however, it
was Conger et al [ l 3 } who first emphasized the possible importance of the relationship. Subsequent reviews
of large numbers of cases of CPM, such as those of
Burcar et a1 [ 111 and Wright et al t471, have shown a
very high incidence of electrolyte derangements, particularly hyponatremia. Although the possibility that
myelinolysis causes the sodium derangements [9, 13,
4 2 ) has been discussed, the view that abnormal sodium levels somehow result in myelinolysis is widely
favored. The clinical data quoted in the introduction
suggested that rapid correction of hyponatremia, rather
than hyponatremia itself, was the critical insult; these
observations stimulated the dog experiments presented
in this and previous reports. A recent paper that documented rapid correction of hyponatremia in relation to
the development of neurological signs in 12 of a series
of 36 patients with CPM [351 stimulated the experiments in rats noted in the introduction. These cumulative clinical and experimental data strongly suggest that
many cases of human myelinolysis are due to the rapid
rise of serum sodium from sustained levels of hyponatremia. In clinical situations, this rapid rise is usually
due to physicians’ therapeutic efforts.
This conclusion confirms the analysis of Aleu and
Terry [4}, who believed that CPM was not only newly
described but was in fact a new disease. They dismissed
the notion that previous generations of pathologists
had overlooked the disease and suggested that some
change in medical practice must have occurred to cause
it. Messert et al [331 speculated that the changes in
medical practice responsible for the emergence of myelinolysis were the impact of the “plastic revolution”
and the development of thiazide diuretics; the latter
possibility had previously been suggested by Cambier
et al [121. These probably are contributory factors.
Perhaps more important was the increasing availability
of serum sodium measurement as a routine clinical test
during the 1950s, an advance in clinical pathology that
allowed more frequent diagnosis of severe hyponatremia. Such enhanced diagnostic capacity logically resulted in therapeutic efforts to correct the sodium derangement. Thus, the stage may have been set for an
increased incidence of myelinolysis.
Although the cause of many cases of human myelinolysis now appears to be clarified, it is possible that
these lesions can occur under circumstances other than
those experimentally investigated to date. It remains
for future studies to show whether myelinolysis develops following a rapid rise of serum sodium from
normal to levels of severe, sustained hypernatremia.
Another variable to be considered is the role of
alcoholism. The overrepresentation of alcoholics in the
myelinolysis population requires explanation. It may
simply relate to the facts that alcoholics can develop
severe hyponatremia and that alcoholics admitted to
the hospital in withdrawal are vigorously treated with
intravenous fluids {7, 8 , 21). Another possibility is that
the alcoholic state or withdrawal of alcohol somehow
makes the brain more sensitive to the rapid rise of
serum sodium from hyponatremia. This is conceivable,
since alcohol reportedly affects brain hydration 171.
The fact that CPM has been described in the setting
of a variety of medical disorders other than alcoholism
[ 191 suggests that seriously ill patients, in general, are
vulnerable to the disease. This fact may simply indicate
that very sick patients face an increased risk of hyponatremia or a greater likelihood of undergoing vigorous correction of that disorder. The other possibility is
that the brains of debilitated patients are less resistant
to a rapid rise of serum sodium. However, carefully
studied cases of CPM are on record in which no serious
underlying illness could be identified [42].
Distribution of Lesions
In the dog, pontine and extrapontine myelinolysis occur together, although in mild cases extrapontine involvement is seen in the absence of a pontine lesion.
Conversely, in humans, the milder cases involve the
pons alone, extrapontine regions being involved only
in the presence of severe pontine lesions. In rat myelinolysis, thalamic and striatal lesions are prominent
but apparently no clear pontine lesion occurs [24, 261.
The neuropathology of metabolic diseases offers precedent for such interspecies variations in regional susceptibility. Presumably, some aspect of the vascular supply, blood-brain barrier, anatomy, or metabolism of the
affected territories makes these regions particularly
susceptible. Although the lesions of CPM are clearly
unsystematized and regional in nature, oligodendroglia
and myelin are more sensitive to the insult than are
neurons and axons. Apparently this regional vulnerability, whatever it is, varies between species.
Certain anatomical features of the pons have been
invoked to explain this regional vulnerability. Okeda
[36] emphasized the close apposition of neurons and
myelinated fibers in the basis pontis. Tomlinson et a1
[42} expanded this observation, stating that the admixture of neurons, oligodendroglia, and myelinated fibers
seen in the pontine basis is also a feature of the extrapontine territories subject to myelinolysis. These authors have suggested that regions of brain with this
histological mixture are for some reason susceptible to
myelinolysis. Messert et a1 [33} emphasized the interlacing longitudinal and transverse fibers of the pontine
basis. They hypothesized that this “grid,” if edematous,
might strangulate the myelinated fibers and thereby
cause myelinolysis. Their postulate follows a line of
previous speculations that edema is important in myelinolysis [23, 31, 38, 41). Goldman and Horoupian
[20) showed a high incidence of myelinolysis in the
lateral geniculate body associated with the pontine disease; they drew an analogy between the lamination of
Laureno: Central Pontine Myelinolysis
239
gray and white matter in the geniculate and the structure of the basis pontis. Kleinschmidt-DeMasters and
Norenberg 124, 261 also drew attention to this admixture of gray and white matter involvement in affected
regions in their rat model of myelinolysis. They proposed that blood-brain barrier breakdown, another frequently suggested etiological mechanism 134, 381, results in edema formation in the richly vascular gray
matter; such edema could then damage the neighboring white matter.
Although the foregoing views are reasonable, they
must be regarded as purely hypothetical for several
reasons. First, the stereotyped symmetrical midline lesion in the pons of the dog is indistinguishable from the
lesion seen in the human pons. However, the pontocerebellar fibers constitute a very thin and ventrally
placed area in the dog pons; this “basis pontis” in the
dog is actually ventral to the lesion and weakens any
argument that attributes pontine vulnerability to the
grid or laminated structure seen in human beings. Second, the pontine lesions in humans frequently extend
into the tegmentum { I ] , even though the latter structure does not share the anatomical features of the basis.
Third, there are regions of the brain in which myelinated fibers are interspersed with layers of gray matter
that do not show myelinolysis. Finally, the factor of
edema, invoked in several theories of myelinolysis, is
little documented. These points must temper enthusiasm for the proposed theories pending further
evidence.
The cerebellar lesions in canine myelinolysis are
maximal in the midline. Cerebellar lesions in rodent
myelinolysis are similarly distributed [251. Because the
cerebellar cortical lesions in so-called alcoholic cerebellar degeneration are also maximal in the midline (anterior-superior vermis), Kleinschmidt-DeMasters and
Norenberg { 2 51 suggested that “alcoholic” cerebellar
degeneration may be due to rapid fluctuations of serum
electrolytes. This hypothesis must contend with several
facts. To begin with, the lesions of the two disorders
are fundamentally different. The lesion in CPM is essentially one of white matter, with relative sparing of
nerve cells, whereas the lesion in “alcoholic” cerebellar
degeneration is neurocellular, particularly affecting the
Purkinje cells 1441. Also noteworthy is the frequent
concurrence of “alcoholic” cerebellar degeneration and
Wernicke’s disease, a disorder resulting from nutritional deficiency-more specifically, from a deficiency
of thiamine. Of 27 patients with Wernicke’s disease in
whom the cerebellum was sectioned sagittally through
the midline, 55.5% showed the typical cerebellar cortical lesions {43]. The probable nutritional cause of
alcoholic cerebellar degeneration is also suggested by
the strong association of this disease with oropharyngeal carcinoma {45].Third, in a series of 1 1 cases of
240 Annals of Neurology
Vol 13 No 3
March 1983
pontine myelinolysis in which the cerebellum in each
case was cut sagittally through the midline, atrophy of
the anterior-superior vermis was observed in only 2
{47]. Furthermore, the most severe forms of human
myelinolysis, those with extensive extrapontine involvement, have never been associated with cortical
degeneration of the anterior-superior vermis {47]. In
summary, except for their prominence in the midline,
the cerebellar lesions of CPM and “alcoholic” cerebellar degeneration are characterized more by their differences than by their similarities.
Marchiafava-Bignami disease, degeneration of the
corpus callosum, resembles CPM in its histological features as well as in its symmetry and central location. For
these reasons the two disorders are often discussed
together. Although canine myelinolysis affects multiple
extrapontine territories, it spares the corpus callosum.
This fact, taken with the absence of corpus callosum
lesions in human cases with extrapontine involvement
[47), tends to negate any relationship between the two
diseases.
Neuropathological Findings in Electrolyte Disorders
Experimental myelinolysis is of special interest because
it expands present-day concepts of the neuropathological changes associated with electrolyte disorders. Although electrolyte derangements may cause serious
disturbances of cerebral function, these disorders have
not been known to result in distinctive structural abnormalities of the nervous system {2, l o ] . Hyponatremia can cause cerebral edema [141 and hypernatremia can lead to cerebral hemorrhage [lb, 171, but
these forms of cerebral damage are consequent to gross
fluid shifts and are in no way specific. The experiments
reported here, however, demonstrate an electrolyte
manipulation resulting in the typical pathological features of a metabolic brain disease, namely, symmetrical
lesions in a stereotyped distribution with characteristic
histological features. In this respect, the occurrence of
myelinolytic lesions following the rapid rise of serum
sodium is unique.
Implications for the Management of Hyponatremia
The fact that rapid correction of sustained hyponatremia can produce pontine myelinolysis should be
weighed during decision making on the management of
hyponatremia. Although severe symptomatic hyponatremia is a serious illness {5}, although hypertonic
saline therapy is a recommended therapy for this disease C30, 391, and even though rapid correction frequently may be achieved without complication [b],the
risk of myelinolysis is real. The experiments reported
here and elsewhere {24, 26, 28, 291 indicate the advantage of moderation in corrective efforts. Particularly in
alcoholics, the rapid normalization of severe, sustained
hyponatremia should be avoided.
Supported by a grant from the Research Foundation of The Washington Hospital Center.
Special thanks to Dr Maurice Victor for his invaluable review of the
manuscript and to Dr Betty Banker for her helpful comments on the
pathological findings. Thanks also to Dr Clive Montalbert for suggestions on methodology, to Mrs Florence Dent and Mrs Nora Fitzpatrick for technical assistance, and to Miss Anne Conway for preparation of the manuscript.
References
1. Adams JH: Central pontine myelinolysis. In Proceedings of the
Fourth International Congress of Neuropathology, 1961,
Munich. Stuttgart, Thieme, 1962, vol 3, pp 303-308
2. Adams R, Victor M: Principles of Neurology. Second edition.
New York, McGraw-Hill, 1981
3. Adams R, Victor M, Mancall E: Central pontine myelinolysis. A
hitherto undescribed disease occurring in alcoholic and nialnourished patients. Arch Neurol 81:154-172, 1959
4. Aleu F, Terry R: Central pontine myelinolysis. A report of two
cases. Arch Pathol 76:140-146, 1963
5 . Arieff AI, Schmidt RW: Fluid and electrolyte disorders and the
central nervous system. In Maxwell M, Kleeman C (eds): Clinical Disorders of Fluid and Electrolyte Metabolism. New York,
McGraw-Hill, 1980, pp 1409-1480
6. Ayus J, Oliver0 J, Frommer J: Rapid correction of severe hyponatremia with intravenous hypertonic saline solution. Am J
Med 72:43-48, 1982
7. Beard J: Body fluid and blood electrolytes. J Pharmacol Exp
Ther 148: 348-355, 1965
8. Beard J, Knott D: The effect of alcohol on fluid and electrolyte
metabolism. In Kissin B, Begleiter H (eds): The Biology of
Alcoholism. New York, Plenum, 1971, vol 1, pp 353--376
9. Berry K, Olszewski J: Central pontine myelinolysis. A case report. Neurology (Minneap) 13:531-537, 1963
10. Blackwood W, Corsellis JAN (eds): Greenfield’s Neuropathology. London, Edward Arnold, 1976
11. Burcar PJ, Norenberg MD, Yarnell P R Hyponatremia and central pontine myelinolysis. Neurology (Minneap) 27:223-226,
1977
12. Cambier J, Masson M, Dairou R, Gray F, Henin D, Laurent D:
Myblinolyse centropontine associee i des lesions plus diffuses de
la substance blanche. Troubles hydroelectrolytiques en relation
avec un traitement diuretique. Rev Neurol (Paris) 133(2):8999, 1977
13. Conger J, McIntyre J, Jacoby W: Central pontine myelinolysis
associated with inappropriate antidiuretic hormone secretion.
Am J Med 47:813-817, 1969
14. Dodge P, Crawford J, Probst J: Studies in experimental water
intoxication. Arch Neurol 5:5 13-529, 1960
15. Espinas OE, Poser CM: Blood hyperosmolality and neurologic
deficit: an experimental study. Arch Neurol 20:182-186, 1969
16. Finberg L Pathogenesis of lesions in the nervous system in hypernatremic stares. 1. Clinical observations in infants. Pediatrics
23~40-45, 1959
17. Finberg L, Luttrell C, Redd H: Pathogenesis of lesions in the
nervous system in hypernatremic states. 11. Experimental studies
of gross anatomic changes and alterations of chemical composition of the tissues. Pediatrics 23:46-53, 1959
18. Finlayson M, Snider S, OlivaLA, Gaulr M: Cerebral and pontine
myelinolysis. Two cases with fluid and electrolyte imbalance and
hypotension. J Neurol Sci 18:399-409, 1973
19. Goebel H H , Herman-Ben Zur P: Central pontine myelinolysis.
In Vinken PJ, Bruyn GW (eds): Handbook of Clinical Neurology. Amsterdam, North Holland, 1976, vol 28: Metabolic and
Deficiency Diseases of the Nervous System, Part 11, pp 285316
20. Goldman JE, Horoupian DS: Demyelinarion of the lateral
geniculate nucleus in central pontine myelinolysis. Ann Neurol
9:185-189, 1981
21. Hilden T, Svendsen TL: Electrolyte disturbances in beer drinkers. Lancet 2:245-246, 1975
22. Kepes J, Reece C, Oxley D: Central pontine myelinolysis in
a 7 year old boy. J Neurol Neurosurg Psychiatry 28:39-47,
1965
23. Klavins JV: Central pontine myelinolysis. J Neuropathol Exp
Neurol 22:302-317, 1963
24. Kleinschmidt-DeMasters BK, Norenberg MD: Rapid correction
of hyponatremia causes demyelination: relation to central pontine myelinolysis. Science 2 11:1068-1070, 1981
25. Kleinschmidt-DeMasters BK, Norenberg MD: Cerebellar degeneration in the rat following rapid correction of hyponatremia.
Ann Neurol 10:561-565, 1981
26. Kleinschmidt-DeMasters BK, Norenberg MD: Neuropathologic observations in electrolyte-induced myelinolysis in the
rat. J Neuropathol Exp Neurol 41:67-80, 1982
27. LandersJW, Chason JL, Samuel VN: Central pontine myelinolysis. A pathogenetic hypothesis. Neurology (Minneap) 15:968971, 1965
28. Laureno R: Experimental pontine and extrapontine myelinolysis.
Trans Am Neurol Assoc 105:354-358, 1980
29. Laureno R Pontine and extrapontine myelinolysis following
rapid correction of experimental hyponatremia. Trans Am
Neurol Assoc 106:98-101, 1981
30. Levinsky NG: Fluids and Electrolytes. In lsselbacher K, Adams
RD, Braunwald E, et al (eds): Harrison’s Principles of Internal
Medicine. Ninth edition. New York, McGraw-Hill, 1980, pp
435-444
3 1. McCormick WF, Danneel CM: Central pontine myelinolysis.
Arch Intern Med 119:444-478, 1967
32. Meier H: Fluids and Electrolytes. In Cornelius CE, Kaneko JJ
(eds): Clinical Biochemistry of Domestic Animals. New York,
Academic, 1963, p 522
33. Messert B, Orrison WW, Hawkins MJ, Quaglieri CE: Central
pontine myelinolysis, considerations on etiology, diagnosis and
treatment. Neurology (NY) 29:147-160, 1979
34. Monteiro L La myelinolyse du centre du pont dans le cadre d’un
nouveau syndrome histopathologique de topographie systematisbe. J Neurol Sci 13:292-314, 1971
35. Norenberg MD, Leslie KO, Robertson AS: Association between rise in serum sodium and central pontine myelinolysis.
Ann Neurol 11:128-135, 1982
36. Okeda R Centrale pontine Myelinolyse. Pathogenetische Aspekte aufgrund morphometrischer Untersuchungen des Bruckenfuses. Acta Neuropathol (Berl) 27:233-246, 1974
37. Paguirigan A, Lefken EB: Central ponrine myelinolysis. Neurology (Minneap) 19:1007-1011, 1969
38. Powers J, McKeever P: Central pontine myelinolysis, an ultrastructural and elemental study. J Neurol Sci 29:65-81,
1976
39. Schwartz WB: Disorders of fluid, electrolyte, and acid base balance. In Beeson P, McDermott W, Wyngaarden J (eds): Cecil’s
Textbook of Medicine. Fifteenth edition. Philadelphia, Saunders, 1979, pp 1950-1969
40. Sherins RJ, Verity MA: Central pontine myelinolysis associated
Laureno: Central Pontine Myelinolysis
24 1
41.
42.
43.
44.
with acute haemorrhagic pancreatitis. J Neurol Neurosurg Psychiatry 31:583-588, 1068
Sima A, Bridvik B: Cenrral pontine myelinolysis, a case report.
Acta Pathol Microbiol Scand [A) 84:73-78, 1976
Tomlinson B, Pierides A, Bradley W: Cenrral pontine myelinolysis. Two cases with associated electrolyte disturbance. Q J
Med [new ser] 45:373-386, 1976
Victor M, Adams RD, Collins GH: The Wernicke-Korsakoff
Syndrome (Contemporary Neurology Series, vol 7). Philadelphia, Davis, 1971
Victor M, Adams RD, Mancall E: A restricted form of cerebellar
242
Annals of N e u r o l o g y
Vol 13 N o 3 March 1983
cortical degeneration occurring in alcoholic patients. AMA Arch
Neurol 1:579-688, 1959
45. Victor M, Laureno R: Neurologic complicarions of alcohol
abuse: epidemiologic aspects. In Schoenberg BS (ed): Neurological Epidemiology: Principles and Clinical Applications. Adv
Neurol 19:603-617, 1978
46. Wakim KG: Predominance of hyponatremia over hypoosmolality in simulation of the dialysis disequilibrium syndrome.
Proc Mayo Clin 44:433-460, 1969
47. Wright DG, Laureno R, Victor M: Pontine and extrapontine
myelinolysis. Brain 102:361-385, 1979
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 111 Кб
Теги
central, rapid, corrections, hyponatremia, pontine, myelinolysis, following
1/--страниц
Пожаловаться на содержимое документа