Neurologic symptoms due to electrolyte disorders are common,occurring in patients with diarrhea, diabetes mellitus, headinjury, renal failure, and many other disorders, especiallyin infants and the elderly. The clinical syndromes of dehydrationand overhydration, often first detected in measurements of plasmasodium or osmolality, are among the most frequent causes ofthe neurologic symptoms, which include irritability, seizures,lethargy, and coma. There are multiple hormonal and neurogenicmechanisms to maintain total body water and the concentrationof solutes (osmolality) within narrow limits. Interruption ofthese homeostatic mechanisms leads to the retention or lossof either water or solute; the result may be overhydration (withhypo-osmolality) or dehydration (with hyperosmolality). Neurologicfunction may be impaired in both hypo-osmolar and hyperosmolarstates. Forty to 60 percent of children with severe hypernatremia(defined as a plasma sodium concentration 160 mmol per liter)have neurologic symptoms, as do a majority of patients withhyponatremia (plasma sodium concentration <120 mmol per liter).The mortality rate in patients with severe hypernatremia isas high as 50 to 60 percent1.
Rapid cerebral dehydration can rupture the blood vessels connectingthe brain to the rigid calvarium. As a protective mechanism,the brain appears to generate new intracellular solute (sometimescalled "idiogenic osmoles")2. Osmoles (or osmolytes) are volume-regulatoryorganic solutes that can accumulate to a high concentrationwithin cells, without adverse effects on cellular structureor function. Intracellular osmolality is thereby increased,minimizing the loss of intracellular brain water. A corollaryof this process is the clinical observation that overly rapidcorrection of hyperosmolar states can be fatal. As extracellularfluid is replaced, the increase in intracellular water associatedwith idiogenic osmoles can lead, it is presumed, to cerebraledema. The rates of accumulation or removal of the idiogenicosmoles are unknown, so that treatment of patients with hyperosmolaland hypo-osmolal states is empirical.
The concept of idiogenic osmoles has recently received supportfrom the identification of several metabolites that are alsoputative organic osmoles -- myo-inositol, N-acetylaspartate,choline (including glycerophosphoryl choline), and taurine --in the brains of laboratory animals3,4. The most important ofthese compounds can be quantified with proton nuclear magneticresonance (NMR) spectroscopy in the brains of humans in vivo5.We describe here a patient with severe dehydration in whom myo-inositoland other substances accumulated and may have functioned asosmolytes, accounting for the presenting neurologic syndromeand slow recovery from hypernatremia.
Methods
Patient
An 18-month-old girl presented with severe dehydration and hypernatremia(plasma sodium concentration, 195 mmol per liter). The childhad not been drinking and was lethargic but otherwise normal.During the correction of dehydration with intravenous glucoseplus 0.45 percent sodium chloride, the child's plasma sodiumconcentration decreased progressively to 144 mmol per literin three days. Her lethargy diminished, but she had two seizures.The rate of fluid replacement was then reduced. The child'scondition improved, she had no further seizures, and she wasalert at discharge after 24 days.
Magnetic resonance imaging (MRI) and proton NMR spectroscopyof the cerebral cortex were performed 4, 7, 12, 22, and 36 daysafter admission to the hospital. The plasma sodium concentrationwas 156 mmol per liter on day 4 and 140 mmol per liter on day36.
The proton NMR spectroscopy and MRI studies were approved bythe local institutional review board, and written informed consentwas obtained from the child's parents and those of normal controlsubjects of similar age.
Proton NMR Spectroscopy
After MRI, quantitative proton NMR spectroscopy was performedin the parietal and occipital regions of the patient's cerebralcortex, containing primarily white matter and gray matter, respectively.Spectra from the same regions in 50 normal infants were alsoacquired. The water content of the brain and the volume of cerebrospinalfluid were evaluated by measuring the large difference in T2transverse relaxation time between them with a quantitativeassay described in detail elsewhere6. Subsequent studies ofthe patient were performed in the same region of the occipitalcortex with use of a single axial MRI sequence to verify thelocation. The images and proton spectra were acquired with aconventional 1.5-T Signa General Electric magnetic resonancescanner.
The first study, performed four days after admission, was obtainedfrom a 12-ml volume of tissue situated across the midline inthe gray matter of the occipital cortex and a region of largelywhite matter in the posterior parietal cortex, with a stimulatedecho sequence, a repetition time of 3.0 seconds (and also of1.5 and 5.0 seconds), an echo time of 30 msec, and a mixingtime of 13.7 msec, with 64 repetitions. Quantitation, performedaccording to the method of Ernst et al.6 and Kreis et al.,7was confirmed by the determination of the transverse relaxationtimes (T1 and T2) of the patient's principal cerebral metabolites.The patient's T1 and T2 relaxation times did not differ fromthose of normal infants8.
Difference spectroscopy, which consists of subtracting the spectraobtained at successive examinations, after scaling them to theirappropriate absolute peak intensities,8 was performed for theidentification of metabolites. The amount of excess organicosmolytes (idiogenic osmoles) was calculated as the total amountof myo-inositol, choline, creatine, glutamine, and N-acetylaspartateless the mean values obtained for these substances in 50 normalinfants8. The excess of each metabolite in the patient couldthen be calculated and expressed in millimoles or milliosmolesper kilogram.
Results
MRI revealed partial holoprosencephaly, with a normal pituitarystalk but no cerebral edema or tumor. Proton NMR spectroscopyof occipital gray matter (Figure 1) and parietal white matter(not shown) revealed several striking abnormalities. The principalabnormality was a reversal of the normal ratio of peak intensitiesbetween the neuronal metabolite N-acetylaspartate and the putativeosmolyte myo-inositol. The ratio of N-acetylaspartate to myo-inositol(normally approximately 2.33)8 was only 0.3 when this infantwas first studied. In addition, the ratios of choline, glutamine(plus glutamate), and scyllo-inositol (or taurine) to creatinewere increased, whereas the ratio of N-acetylaspartate to creatinewas reduced. The relaxation values for the metabolites, whichif altered by disease can contribute to differences in spectra,did not differ from those in normal infants. The explanationfor the changing metabolite ratios became clear after quantitativeprocedures involving proton NMR spectroscopy were applied6,7.The principal change was in the concentration of myo-inositol,which was three times normal, whereas that of N-acetylaspartatewas normal or only minimally reduced as compared with the concentrationin an age-matched normal subject. Concentrations of cholineand glutamine (plus glutamate) were also increased.
Figure 1. Short-Echo-Time Proton NMR Spectrum of Cortical Gray Matter in an Infant with Severe Dehydration.
The striking differences from normal in this patient were the apparent increase in myo-inositol and the decrease in N-acetylaspartate. The increase in myo-inositol was accompanied by increases in the concentrations of its stereoisomer, scyllo-inositol (sI),9,10,11 and choline-containing compounds (Cho), especially glycerophosphoryl choline (GPC), glutamine plus glutamate (Glx), and creatine.
The sequential spectra obtained during correction of the dehydrationare shown in Figure 2. The myo-inositol concentration fell progressively,as did those of choline and creatine. The concentration of N-acetylaspartateincreased slightly, then returned to normal. The negative peaksin the difference spectra indicate metabolites that were lowerin concentration after treatment. The water content of the brain(84 percent on day 4 and 82 percent on day 12) and the T2 relaxationtime for water were normal for age8.
Figure 2. Changes in the Concentrations of Principal Cerebral Organic Osmolytes during the Correction of Severe Dehydration.
All spectra were obtained from the same location in the occipital gray matter under identical conditions. Spectra were processed and scaled identically to permit the subtraction of sequential spectra by difference spectroscopy. The differences identify more clearly the progressively declining concentrations of several metabolites as negative peaks (labeled for days 7 and 12), with a possible elevation in the resonance peak assigned to the neuronal marker N-acetylaspartate (NAA). Cho denotes choline, mI myo-inositol, and Cr creatine.
The calculated concentrations of the five principal metabolitesshown in spectra of the infant's brain (myo-inositol, choline,creatine, glutamine, and N-acetylaspartate) on days 4 to 36are shown in Table 1. The concentration of myo-inositol on day4 was three times higher than that in normal subjects of thesame age. Other substances with elevated concentrations at thattime of measurement were choline, scyllo-inositol, creatine,and glutamine (plus glutamate). These levels all decreased tonormal during the 36-day study period. If we assume that myo-inositoland the other organic molecules are the same osmolytes identifiedin earlier studies in animals (with the exception that scyllo-inositolis found in the human brain, rather than taurine, which is foundin rat brain), then the excess intracellular osmotic pressurein the brain can be estimated, both when the child was dehydratedand obtunded and later, after correction. We estimate an excessof 17 mOsm initially, falling to about 6 mOsm after 7 days,and returning to normal only after 36 days. The change in totalbrain water was not significant (the standard error associatedwith the method was ±3 percent).
Table 1. Results of Clinical Proton NMR Spectroscopy for Cerebral Osmometry of the Brain of an Infant with Hypernatremia and Normal Subjects.
Discussion
Holoprosencephaly is commonly associated with a deficiency ofthirst and with dehydration,2,12 leading to hypernatremia andneurologic symptoms. These changes were associated with increasedconcentrations of osmotically active solute in the brain. Wehave also found changes in these metabolites in adults13 andchildren recovering from diabetic ketoacidosis and in patientswith hypernatremia and hyponatremia14.
The direct determination of patterns of disordered cerebralorganic osmolytes by proton NMR spectroscopy may be valuablein guiding therapy. In the presence of severe changes in intracellularosmolytes in the brain (as in this infant), a much slower correctionof the plasma sodium concentration, possibly over a period of7 to 10 days, may be indicated. Conversely, a normal patternof cerebral osmolytes may permit the safe and rapid replacementof fluids. Repeating proton NMR spectroscopy after 7 to 10 days,or earlier in the event of seizures, may show that a normalcerebral-osmolyte profile has yet to be achieved and may leadto even slower replacement therapy. Central pontine myelinolysis,a rare but fatal complication of severe hypernatremia, may becomepredictable and it may be possible to prevent it.
Disorders of cerebral osmotic regulation may be more prevalentthan has been hitherto assumed. A simplified procedure for NMRspectroscopy of the brain, automated single-voxel acquisitionproton NMR spectroscopy (Probe, General Electric Medical Systems),15now brings this means of measuring osmolyte accumulation withinthe scope of any hospital with an MRI scanner of 1.5-T fieldstrength.
Supported by a grant from the L.K. Whittier Foundation, SouthPasadena, Calif.
We are indebted to Linda Fisher, M.D., for advice on the medicaltreatment of this child and to Richard Yadley, M.D., for readingthe MRI scans.
Source Information
From the Magnetic Resonance Spectroscopy Unit, Huntington Medical Research Institutes (J.H.L., B.D.R.); and Huntington Memorial Hospital (E.A.) -- both in Pasadena, Calif.
Address reprint requests to Dr. Ross at Huntington Medical Research Institutes, 660 S. Fair Oaks Ave., Pasadena, CA 91105.
References
Fraser CL, Arieff AI. Metabolic encephalopathy associated with water, electrolyte and acid-base disorders. In: Maxwell MH, Kleeman CR, Narins RG, eds. Clinical disorders of fluid and electrolyte metabolism. New York: McGraw-Hill, 1987:1153-69.
Arieff AI, Kleeman CR. Studies on mechanisms of cerebral edema in diabetic comas: effects of hyperglycemia and rapid lowering plasma glucose in normal rabbits. J Clin Invest 1973;52:571-583.
Gullans SR, Verbalis JG. Control of brain volume during hyperosmolar and hypo-osmolar conditions. Annu Rev Med 1993;44:289-301. [CrossRef][Medline]
Lien Y-HH, Shapiro JI, Chan L. Effects of hypernatremia on organic brain osmoles. J Clin Invest 1990;85:1427-1435.
Ross BD, ed. Proton spectroscopy in clinical medicine. NMR Biomed 1991;4:47-116. [Medline]
Ernst T, Kreis R, Ross BD. Absolute quantitation of water and metabolites in the human brain. Part I: compartments and water. J Magn Reson 1993;102:1-8. [CrossRef]
Kreis R, Ernst T, Ross BD. Absolute quantitation of water and metabolites in the human brain. Part II: metabolite concentrations. J Magn Reson 1993;102:9-19. [CrossRef]
Kreis R, Ernst T, Ross BD. Development of the human brain: in vivo quantification of metabolite and water content with proton magnetic resonance spectroscopy. Magn Reson Med 1993;30:424-437. [Medline]
Michaelis T, Moats RA, Lien Y-HH, Ross BD. Depletion of cerebral inositols in rat and human: a HPLC and 1H MRS study of an animal model. In: Proceedings: 12th meeting of the Society of Magnetic Resonance in Medicine, New York, August 14-20, 1993. Berkeley, Calif.: Society of Magnetic Resonance in Medicine, 1993:437. abstract.
Moats RA, Lien Y-HH, Filippi D, Ross BD. Decrease in cerebral inositols in rats and humans. Biochem J 1993;295:15-18.
Lien Y-HH, Michaelis T, Moats RA, Ross BD. Scyllo-inositol depletion in hepatic encephalopathy. Life Sci 1994;54:1507-1512. [Medline]
Morrison G, Singer I. Hyperosmolar states. In: Maxwell MH, Kleeman CR, Narins RG, eds. Clinical disorders of fluid and electrolyte metabolism. New York: McGraw-Hill, 1987:481-518.
Kreis R, Ross BD. Cerebral metabolic disturbances in patients with subacute and chronic diabetes mellitus: detection with proton MR spectroscopy. Radiology 1992;184:123-130. [Free Full Text]
Videen J, Michaelis T, Raj AS, Pinto P, Ross BD. Moderate hyponatremia modifies brain metabolism. In: Proceedings: 2nd meeting of the Society of Magnetic Resonance, San Francisco, August 8-12, 1994. Berkeley, Calif.: Society of Magnetic Resonance, 1994:193. abstract.
Webb PG, Sailasuta N, Kohler SJ, Raidy T, Moats RA, Hurd RE. Automated single-voxel proton MRS: technical development and multisite verification. Magn Reson Med 1994;31:365-373. [Medline]
Koul, P. B.
(2009). Diabetic Ketoacidosis: A Current Appraisal of Pathophysiology and Management. CLIN PEDIATR
48: 135-144
[Abstract]
Helmy, A., Vizcaychipi, M., Gupta, A. K.
(2007). Traumatic brain injury: intensive care management. Br J Anaesth
99: 32-42
[Abstract][Full Text]
Minati, L., Grisoli, M., Bruzzone, M. G.
(2007). MR Spectroscopy, Functional MRI, and Diffusion-Tensor Imaging in the Aging Brain: A Conceptual Review. J Geriatr Psychiatry Neurol
20: 3-21
[Abstract]
Bartsch, A. J., Homola, G., Biller, A., Smith, S. M., Weijers, H.-G., Wiesbeck, G. A., Jenkinson, M., Stefano, N. D., Solymosi, L., Bendszus, M.
(2007). Manifestations of early brain recovery associated with abstinence from alcoholism. Brain
130: 36-47
[Abstract][Full Text]
Friedman, S. D., Shaw, D. W. W., Artru, A. A., Dawson, G., Petropoulos, H., Dager, S. R.
(2006). Gray and White Matter Brain Chemistry in Young Children With Autism.. Arch Gen Psychiatry
63: 786-794
[Abstract][Full Text]
Righini, A., Ramenghi, L., Zirpoli, S., Mosca, F., Triulzi, F.
(2005). Brain Apparent Diffusion Coefficient Decrease During Correction of Severe Hypernatremic Dehydration. Am. J. Neuroradiol.
26: 1690-1694
[Abstract][Full Text]
Yoon, S. J., Lee, J. H., Kim, S. T., Chun, M. H.
(2005). Evaluation of traumatic brain injured patients in correlation with functional status by localized 1H-MR spectroscopy. Clin Rehabil
19: 209-215
[Abstract]
Brockmann, K., Dechent, P., Wilken, B., Rusch, O., Frahm, J., Hanefeld, F.
(2003). Proton MRS profile of cerebral metabolic abnormalities in Krabbe disease. Neurology
60: 819-825
[Abstract][Full Text]
Inward, C D, Chambers, T L, Edge, J
(2002). Fluid management in diabetic ketoacidosis. Arch. Dis. Child.
86: 443-444
[Full Text]
Oddie, S, Richmond, S, Coulthard, M
(2001). Hypernatraemic dehydration and breast feeding: a population study. Arch. Dis. Child.
85: 318-320
[Abstract][Full Text]
Axford, J S, Howe, F A, Heron, C, Griffiths, J R
(2001). Sensitivity of quantitative 1H magnetic resonance spectroscopy of the brain in detecting early neuronal damage in systemic lupus erythematosus. Ann Rheum Dis
60: 106-111
[Abstract][Full Text]
Sinson, G., Bagley, L. J., Cecil, K. M., Torchia, M., McGowan, J. C., Lenkinski, R. E., McIntosh, T. K., Grossman, R. I.
(2001). Magnetization Transfer Imaging and Proton MR Spectroscopy in the Evaluation of Axonal Injury: Correlation with Clinical Outcome after Traumatic Brain Injury. Am. J. Neuroradiol.
22: 143-151
[Abstract][Full Text]
Kantarci, K., Jack, C. R. Jr., Xu, Y. C., Campeau, N. G., O'Brien, P. C., Smith, G. E., Ivnik, R. J., Boeve, B. F., Kokmen, E., Tangalos, E. G., Petersen, R. C.
(2000). Regional metabolic patterns in mild cognitive impairment and Alzheimer's disease: A 1H MRS study. Neurology
55: 210-217
[Abstract][Full Text]
Muir, A.
(2000). Cerebral Edema in Diabetic Ketoacidosis: A Look Beyond Rehydration. J. Clin. Endocrinol. Metab.
85: 509-513
[Full Text]
McManus, M. L., Churchwell, K. B., Strange, K.
(1995). Regulation of Cell Volume in Health and Disease. NEJM
333: 1260-1267
[Full Text]
Schulman, M., Lee, J. H., Arcinue, E., Ross, B. D.
(1994). Organic Osmolytes in the Brain of an Infant with Hypernatremia. NEJM
331: 1776-1777
[Full Text]
Molteni, K. H.
(1994). Initial Management of Hypernatremic Dehydration in the Breastfed Infant. CLIN PEDIATR
33: 731-740
160 mmol per liter) have neurologic symptoms, as do a majority of patients with hyponatremia (plasma sodium concentration <120 mmol per liter). The mortality rate in patients with severe hypernatremia is as high as 50 to 60 percent1. 
Previous
