Authors' reply



As an anaesthetist involved in the management of women with hypertensive disease in pregnancy, I read with interest the recent review by Idama and Lindow (Vol. 105, March 1998)1 on the clinical pharmacology of magnesium sulphate in obstetrics.

It is possible that the relative reluctance of British clinicians to use magnesium for the treatment of eclampsia originated from the apparently widely held, but quite erroneous notion, that exogenously administered magnesium does not gain access to the central nervous system and therefore could not exert central anticonvulsant activity. This notion may have arisen from an erroneous interpretation of early studies on humans and animals. The mistaken concept that magnesium must act only as a peripheral neuromuscular blocking agent in eclampsia was thereby favoured.

Cohen in 19272 first observed that in human subjects CSF magnesium levels failed to reflect fluctuations in plasma levels. Other early workers confirmed this fact. Over 30 years later Oppelt et al.3 also demonstrated that if non-radioactive magnesium is infused intravenously in dogs, the concentration of magnesium in CSF remains unchanged. This is because CSF magnesium levels are strictly controlled by rigorous homeostatic mechanisms. If however radioactive 28Mg2+ was infused intravenously there was a rise in CSF 28Mg2+ levels and a corresponding fall in plasma 28Mg2+ concentration clearly indicating exchange between plasma magnesium and CSF magnesium. It is important to emphasise therefore that there is sound experimental evidence for the passage of magnesium from plasma into the CSF and brain. As all magnesium in the body originates from dietary sources, brain and CSF magnesium must necessarily pass into the central nervous system as a matter of routine. Clearly, this physiological process occurs with an intact blood-brain and blood-CSF barrier.

It is not necessary for the blood-brain barrier to be in anyway disrupted for exogenously administered magnesium to gain access to the central nervous system. This presumably is why magnesium can act prophylactically to reduce the frequency of eclampsia in women with severe pre-eclampsia as reported by Coetzee et al.4

Therefore, it is not the case that magnesium only gains access to the brain and CSF in eclamptics because fitting has disrupted the blood brain barrier. The blood-brain barrier may well be normal in the majority of women with eclampsia. One may conjecture that the relatively specific anticonvulsant action of Mg2+ in eclampsia is due to a combination of cerebral arteriolar vasodilatation and N-methyl-D-aspartate receptor blockade, allowing Mg2+ to interrupt explosive nitric oxide formation secondary to focal or global cerebral ischaemia.