Neuroscience & Medicine, 2010, 1, 60-63
doi:10.4236/nm.2010.12009 Published Online December 2010 (http://www.SciRP.org/journal/nm)
Cerebrospinal Fluid Magnesium Level in Different
Neurological Disorders
Haensch Carl-Albrecht
Department of Neurology, University of Witten/Herdecke, HELIOS Klinikum Wuppertal, Germany
Received August 30th, 2010; revised September 19th, 2010; accepted September 26th, 2010
ABSTRACT
Magnesium (Mg) is an essential cofactor for many en zymatic reaction s, esp ecia lly those in vo lved in en ergy metabolism.
The aim of the present study was to determine the CSF concentration of Mg in various neurological disorders (n = 72)
and in healthy subjects (n = 75). The control group included 35 males and 40 females, aged 16-89 years (mean age 53
years) who were subjected to a lumbar puncture for diagnostic reasons. The CSF examination was normal mainly as
concerns the macroscopically examination, the leukocyte count and the protein level. The determination of Mg was
performed with xylidyl-blue photometry. Our normal CSF Mg mean value was 0.97 ± 0.08 mmol/l (range 0.6-1.4
mmol/l). In the group of patients (n = 11) with convulsive seizures a slightly but significantly lower Mg were revealed
(0.92 ± 0.03 mmol/l; p = 0.001; paired two-tailed S tuden t’s t-tests). No statistically significant change of CSF Mg levels
was noted in patients su ffering from alcohol withdrawal syndrome, multiple sclerosis or Bell’s palsy. Our results indi-
cate that magnesium deficiency may play a role for seizure manifestation even in patients with a moderate low Mg
without neurological signs. Low CSF magnesium is associated with epilepsy, further studies may determine the influ-
ence of anti-epileptic drug therapy on CSF magnesium levels.
Keywords: Magnesium, Cerebrospinal Fluid, Seizures, Epilepsy
1. Introduction
The significance of trace elements concentration in the
serum and various body fluids has been demonstrated in
a number of publications. However, magnesium deter-
minations in the cerebrospinal fluid (CSF) in neurologi-
cal diseases has been reported only in a small number of
communications [1-3]. Recently, a greater emphasis has
been given to the role of trace elements in the function of
the nervous syste m both in normal and pathological con-
ditions. However, the results are variable and the object
remains open for further investigation. It has been sug-
gested that low serum Mg has occasionally been associ-
ated with significant effects on the central nervous sys-
tem especially in epilepsy and alcoholism. Magnesium
deficiency-dependent seizures may be a relevant factor
for epileptogenesis. Further studies in the future may
determine the influence of anti-epileptic drug therapy on
CSF magnesium levels. The existing knowledge about
the impact of magnesium on epilepsy and other neuro-
logical diseases is poor and controversial. Therefore, in
this study, we purposed evaluation of the level of CSF
concentration of Mg in a total of seventy-two neurologi-
cal patients and seventy-five healthy controls.
2. Methods
2.1. Patients
We studied prospectively all patients over 16 years who
were subjected to a lumbar puncture for diagnostic rea-
sons. The control group (n = 75) had mainly attended the
Neurological Department because of headache or psycho-
neurosis and the CSF examination was normal as con-
cerns the mac roscopica lly examination, the leuko cyte c o unt
and the protein level. Patients who were treated with
magnesium or calcium supplements or any drugs that
contain these electrolytes were excluded. A contraindica-
tion for lumbar puncture or a loss of indication for per-
formance was also exclusion criteria. Every patient had a
detailed history taken and clinical ex amination, including
general and neurological status by a neurologist. The lum-
bar puncture was mainly performed on admission. CSF
and serum magnesium levels were checked immediately
afterwards.
2.2. Chemical Analysis
The determination of Mg was performed with xylidyl-
blue photometry (Hitachi 717). CSF analysis included:

Cerebrospinal Fluid Magnesium Level in Different Neurological Disorders 61
cell count (normal values: total white cells < 5 cells/ mm3,
protein (normal value < 0.5 g/l), glucose (normal value >
50% of blood glucose) and lactate (normal value 1.1 to
3.2 mmol/l).
2.3. Statistics
Paired two-tailed Student’s t-tests were used to compare
CSF magnesium in different groups. The study was ap-
proved by the ethics committee of th e University of Wit-
ten/Herdecke according to the revised Declaration of
Helsinki.
3. Results
Seventy-two patients (40 male and 32 female) and sev-
enty-five healthy subjects (35 males and 40 females)
fulfilled the criteria of the study. Their ages ranged from
16 to 93 years (mean 53.6 y). The patient-group included
neurological diagno sis like bell’s palsy (n = 12), multiple
sclerosis (n = 10), alcohol withdraw al syndro me (n = 16) ,
Lyme disease (n = 4), stroke (n = 6), HSV-encephalitis (n
= 3), meningitis (n = 3), dementia (n = 3), polyneuropa-
thy (n = 5), Guillain Barrè syndrome (n = 2) and other (n
= 8). The magnesium CSF level averaged 0.97 mmol/l,
SD 0.08 mmol/l (range 0.6-1.4 mmol/l; normal values
0.77-1.17 mmo l/l). Serum Mg and CSF Mg do not corre-
lated. CSF magnesium varied not as a function of age
(Figure 1). The correlation between CSF Mg and CSF
protein failed to reach significance (Figure 2). In the
group of patients with convulsive seizures (n = 11) a
slightly but significantly lower Mg were revealed (0 .92 ±
0.03 mmol/l; p = 0.001; paired two-tailed Student’s
t-tests; Figure 3). No statistically significant change of
CSF Mg levels was noted in patients suffering from al-
cohol withdrawal syndrome, multiple sclerosis or Bell’s
palsy.
4. Discussion
There are many previous studies of serum Mg concentra-
tion of normals and patients, but only few of CSF con-
centration of the same metal. Our normal values are in
accordance with these [4-6].
Hypomagnesemia is common in alcoholic patients
admitted to the hospital; in one study for example, the
Figure 1. Age distribution of CSF-magnesium.
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
0 102030405060708090100
Age [years]
CSF-Mg [mmol/l]

Cerebrospinal Fluid Magnesium Level in Different Neurological Disorders 62
Figure 2. Correlation betw e e n CSF-Mg and CSF-protein.
Figure 3. Significantly lower CSF-Mg in the group of pa-
tients with convulsive seizures.
prevalence was 30 percent [7]. In a prospective study of
82 patients with alcohol-related admission diagnoses 14
were hypomagnesemic while 16 had a normal plasma
magnesium concentration [8]. However, consistent with
other studies we found normal CSF magnesium in alco-
hol withdrawal syndrome and a lack of correlation be-
tween plasma and CSF magnesium levels [9]. Although
the role of magnesium in neuronal function is not com-
pletely understood, a lowering of CSF or brain magne-
sium can induce epileptiform activity and there is an as-
sociation between decreased CSF magnesium and the
development of seizures [10-15]. Over 300 intracellular
enzyme processes depend on magnesium for activation.
Many of these enzymes, including pyruvate kinase, ade-
nylate cyclase, and pyrophosphatase, undergo a confor-
mational change when magnesium binds, thus activating
the enzyme. Altering the magnesium concentrations within
the physiological range results in concentration-dependent
changes in the Na+/K+ -ATPase activity, which may re-
sult in increased seizure susceptibility [11]. Within the
cell, an optimal concentration of magnesium is required
for DNA synthesis, making protein synthesis in part de-
pendent on adequate magnesium levels [16]. In addition,
magnesium is required for cell membrane fluidity and
stability and regulates the permeability of membrane to
other cations such as potassium and calcium. Finally,
magnesium regulates the binding or function of opiate and
N-methyl-D-aspartate (NMDA) receptors. Magnesium, at
physiological concentrations, blocks NMDA receptors in
neurones [17]. Many NMDA receptor antagonists have a
potent antiepileptic properties and activation of NMDA
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
0100 200 300 400 500 600 700
CSF-protein [mg/dl]
CSF-Mg [mmol/l]

Cerebrospinal Fluid Magnesium Level in Different Neurological Disorders 63
receptors may also contribute to epileptogenesis. The
mechanisms of epileptogenesis are not well established.
Several studies in the last few years suggested that the
body electrolytes, level of some trace elements, and mem-
brane lipid peroxidation due to increase in free radicals or
decrease in activities of antioxidant defense mechanisms
may be causally involv ed in some forms of ep ileps ies and
also to increase the recurrence of seizures. The therapeu-
tic consequence of these findings may have some impor-
tance.
5. Conclusions
Our results indicate that magnesium deficiency may play
a role for seizure manifestation even in patients with a
moderate low CSF Mg without neurological signs. In
accordance with our results Miyamoto et al. showed re-
cently in a study in convulsive children significantly
lower CSF Mg than in non-convu lsive children [18]. The
role of low CSF magnesium concentration in the induc-
tion of seizures is not fully understood but may involve
alteration of NMDA receptor regulation in the central
nervous system.
REFERENCES
[1] B. Bocca, A. Alimonti, O. Senofonte, A. Pino, N.
Violante, F. Petrucci, G. Sancesario and G. Forte, “Metal
Changes in CSF and Peripheral Compartments of Parkin-
sonian Patients,” Journal of the Neurological Sciences,
Vol. 248, No. 1-2, 2006, pp. 23-30.
doi:10.1016/j.jns.2006.05.007
[2] E. Kapaki, J. Segditsa and C. Papageorgiou, “Zinc, Cop-
per and Magnesium Concentration in Serum and CSF of
Patients with Neurological Disorders,” Acta Neurologica
Scandinavica, Vol. 79, No. 5, 1989, pp. 373-378.
doi:10.1111/j.1600-0404.1989.tb03803.x
[3] H. McConnell and J. Bianchine, “Cerebrospinal Fluid in
Neurology and Psychiatry,” Chapmann & Hall, London,
1994, pp. 1-322.
[4] A. Bayir, A. Ak, H. Kara and T. K. Sahin, “Serum and
Cerebrospinal Fluid Magnesium Levels, Glasgow Coma
Scores, and in-Hospital Mortality in Patients with Acute
Stroke,” Biological Trace Element Research, Vol. 130,
No. 1, 2009, pp. 7-12. doi:10.1007/s12011-009-8318-9
[5] L. Gerhardsson, T. Lundh, L. Minthon and E. Londos,
“Metal Concentrations in Plasma and Cerebrospinal Fluid
in Patients with Alzheimer's Disease,” Dementia and
Geriatric Cognitive Disorders, Vol. 25, No. 6, 2008, pp.
508-515. doi:10.1159/000129365
[6] M. G. Harrington, A. N. Fonteh, R. P. Cowan, K. Perrine,
J. M. Pogoda, R. G. Biringer and A. F. Huhmer, “Cere-
brospinal Fluid Sodium Increases in Migraine,” Headache,
Vol. 46, No. 7, 2006, pp. 1128-1135.
doi:10.1111/j.1526-4610.2006.00445.x
[7] M. Elisaf, M. Merkouropoulos, E. V. Tsianos and K. C.
Siamopoulos, “Pathogenetic Mechanisms of Hypomag-
nesemia in Alcoholic Patients,” Journal of Trace Ele-
ments in Medicine and Biology, Vol. 9, No. 4, 1995, pp.
210-214.
[8] E. Ryzen, T. A. Nelson and R. K. Rude, “Low Blood
Mono nuclear Cell Magnesium Conte nt and Hypocalcemia in
Normomagnesemic Patients,” Western Journal of Medi-
cine, Vol. 147, No. 5, 1987, pp. 549-553.
[9] P. Kramp, R. Hemmingsen and O. J. Rafaelsen, “Magnesium
Concentrations in Blood and Cerebrospinal Fluid during De-
lirium Tremens,” Psychia try Resear ch, Vol. 1, No. 2 , 1979, pp.
161-171. doi:10.1016/0165-1781(79)90057-X
[10] P. A. Abdelmalik, W. M. Burnham and P. L. Carlen, “In-
creased Seizure Susceptibility of the Hippocampus Com-
pared with the Neocortex of the Immature Mouse Brain in
Vitro,” Epilepsia, Vol. 46, No. 3, 2005, pp. 356-366.
doi:10.1111/j.0013-9580.2005.34204.x
[11] M. Derchansky, E. Shahar, R. A. Wennberg, M. Sam oi l ov a,
S. S. Jahromi, P. A. Abdelmalik, L. Zhang and P. L. Car-
len, “Model of Frequent, Recurrent, and Spontaneous
Seizures in the Intact Mouse Hippocampus,” Hippocam-
pus, Vol. 14, No. 8, 2004, pp. 935-947.
doi:10.1002/hipo.20007
[12] E. Shahar, M. Derchansky and P. L. Carlen, “The Role
of Altered Tissue Osmolality on the Characteristics and
Propagation of Seizure Activity in the Intact Isolated
Mouse Hippocampus,” Clinical Neurophysiology, Vol.
120, No. 4, 2009, pp. 67 3-678.
doi:10.1016/j.clinph.2009.01.014
[13] J. Solger, U. Heinemann and J. Behr, “Electrical and
Chemical Long-Term Depression Do Not Attenuate Low-
Mg2+-Induced Epileptiform Activity in the Entorhinal
Cortex,” Epilepsia, Vol. 46, No. 4, 2005, pp. 509-516.
doi:10.1111/j.0013-9580.2005.41204.x
[14] A. K. Sood, R. Handa, R. C. Malhotra and B. S. Gupta, “Se-
rum, CSF, RBC & Urinary Levels of Magnesium & Calcium
in Idiopathic Generalised Tonic Clonic Seizures,” Indian
Journal of Medical Resesrch, Vol. 9 8, 1993, pp. 152-154.
[15] L. J. Voss and J. W. Sleigh, “Stability of Brain Neocorti-
cal Slice Seizure-Like Activity during Low-Magnesium
Exposure: Measurement and Effect of Artificial Cerebro-
spinal Fluid Temperature,” Journal of Neuroscience Methods,
2010, [Epub ahead of print], 10.1016/j. jneumeth.2010.07.
025.
[16] M. S. George, D. Rosenstein, D. R. Rubinow, M. A.
Kling and R. M. Post, “CSF Magnesium in Affective Dis-
order: Lack of Correlation with Clinical Course of Treat-
ment,” Psychiatry Research, Vol. 51, No. 2, 1994, pp.
139-146. doi:10.1016/0165-1781(94)90033-7
[17] M. E. Morris, “Brain and CSF Magnesium Concentrations
during Magnesium Deficit in Animals and Humans: Neu-
rological Symptoms,” Magnesium Research, Vol. 5, No. 4,
1992, pp. 303-313.
[18] Y. Miyamoto, H. Yamamoto, H. Murakami, N. Kamiyama
and M. Fukuda, “Studies on Cerebrospinal Fluid Ionized
Calcium and Magnesium Concentrations in Convulsive
Children,” Pediatrics International, Vol. 46, No. 4, 2004,
pp. 394-397. doi:10.1111/j.1442-200x.2004.01922.x