Severe Muscle Cramp due to Acute
Hypomagnesaemia in Haemodialysis
British Medical Journal, 1969, 2, 804-805
Attention has recently been drawn to the complications of hypermagnesaemia induced by haemodialysis for chronic renal
failure (Govan et al., 1968).
We report here a case of severe muscle cramp due to acute hypomagnesaemia occurring in chronic intermittent haemodialysis. Wacker and Parisi (1968)
extensively reviewed abnormalities of magnesium metabolism in man; there is no reported evidence for acute hypomagnesaemia related to dialysis.
The patient, a 38-year-old white South African with advanced chronic glomerulonephritis, was started on regular twice-weekly
haemodialysis early in 1968, a Kolff kidney with twin coil 145 being used for nine hours per dialysis. After about four months
he began to develop severe muscle cramps in both legs and to a lesser degree in both arms. These involved the smaller muscles in
particular and usually came on about six to seven hours after the start of dialysis, gradually fading in the 12 hours after the end of
dialysis. The blood pressure and pulse remained normal throughout, and no excessive loss of weight was noted. Both oral quinine
and intravenous saline were given without any effect on the cramp.
Physical examination failed to show any abnormal signs beyond the muscle spasm. Both Chvostek and Trousseau signs were
negative. Plasma calcium and potassium were normal both at the beginning and at the end of dialysis. Predialysis magnesium, however,
was 2’5 mg/100 ml. (normal range 1-7-21 mg./100 ml.), and this fell to 1P3 mg./100 ml. at the end of dialysis (see Table).
Magnesium Concentrations (mg./100 ml.)
Predialysis Mg 2-5 2-5 1-3-2-2
Postdialysis Mg .. 1-3 1-8 1 0-1 9
Fall in plasma Mg concentration
during dialysis 1-2 0-7 0-10-7
Bath-water Mg. .. 0-8 1-8 0-8
No magnesium was added to the dialysing fluid, and the bath magnesium concentration was normally only 0-8 mg./100 ml.
During one of the acute cramps 4 mEq of magnesium in the form of 10% MgSO4 was administered intravenously, This brought
about a sudden and dramatic relief of the cramp, which did not recur during that dialysis. At the beginning of the next dialysis the
plasma magnesium was again 2’5 mg./100 ml. Severe cramp recurred during dialysis and was again relieved by a similar dose
of magnesium sulphate.
Magnesium sulphate was next added to the bath-water in order to bring the magnesium concentration up to 1-8 mg./100 ml. Since
then the patient has had no severe muscle cramp.
Although all of our patients on chronic intermittent haemodialysis are dialysed against a bath-water magnesium concentration
of 0.8 mg/100 ml., we have not seen these excruciating cramps in any others. The predialysis plasma magnesium
concentrations in six patients ranged from 1 3 to 2-2 mg/100 ml. (see Table).
Serial predialysis estimations over several months showed no tendency for hypomagnesaemia to develop.
Nevertheless, in none of these patients was the plasma magnesium level as high as 2-5 mg./100 ml. at the beginning
of dialysis, but the postdialysis magnesium levels (10 to 19 mg./100 ml.) are comparable with those found in the above patient.
At the time of the onset of severe cramps the patient’s plasma magnesium was 16 mg./100 ml., which is only just below the
normal range. This suggests that the clinical symptoms were not related to the absolute level of plasma magnesium but
rather to the rate and extent of fall in plasma magnesium level during dialysis. Symptoms occurred in the only patient with
a fall in plasma magnesium concentration exceeding 1 mg/100 ml. during the period of dialysis, and he became symptomfree
as soon as this rate of decline was reduced.
This situation may be related to the fact that this patient had a rather higher than normal plasma magnesium level. This
is an unusual feature in patients on long-term chronic intermittent dialysis, though hypermagnesaemia has been described
in acute renal failure with the use of flame photometry estimation (Wacker and Vallee, 1957).
We have not added magnesium sulphate to the bath-water of our other patients on chronic dialysis, since none of them
appear to have had any clinical evidence of magnesium deficiency.
Our plasma magnesium estimations were done by a colorimetric method (Bohuon, 1962). Although objections have been
raised to this method, and the atomic absorptiometer is superior, the findings in this case are probably valid because (1) the
estimations were repeatable, and (2) the important features in this case were the relative values before, during, and after
dialysis rather than the absolute values themselves.
D. R. TRIGER, M.A., B.M., B.CH.,
Resident Medical Officer.
A. M. JOEKES, M.A., B.M., F.R.C.P.,
St. Peter’s, St. Paul’s, and St. Philip’s Hospitals, London W.C.2.
Bohuon, G. (1962). Clinica Chimica Acta, 7, 811.
Govan, J. R., Porter, C. A., Cook, J. G. H., Dixon, B., and Trafford,
J. A. P. (1968). British Medical Yournal, 2, 278.
Wacker, W. E. C., and Panisi, A. F. (1968). New England 7ournal of
Medicine, 278, 658.
Wacker, W. E. C., and Vallee, B. L. (1957). New England Yournal of
Medicine, 257, 1254.