Article Text

Download PDFPDF

Central pontine myelinolysis temporally related to hypophosphataemia
  1. A W Michell,
  2. D J Burn,
  3. P J Reading
  1. Regional Neurosciences Centre, Newcastle-upon-Tyne, UK
  1. Correspondence to:
 Dr Michell; 

Statistics from

Central pontine myelinolysis (CPM) is known to be associated with the rapid correction of severe hyponatraemia. However, there have been case reports of CPM occurring in normonatraemic patients.1 Here we describe two patients in whom chronic alcohol abuse led to profound hypophosphataemia that was closely temporally related to the development of CPM.

Case 1

A 29 year old woman was admitted for investigation of painless jaundice of 10 days’ duration. She had consumed 100–140 units of alcohol a week for the preceding 18 months and had been noted to have mildly deranged serum transaminase levels one year previously.

On admission she was fully oriented with normal speech and gait. She had a mild postural tremor but no asterixis. A plasma biochemical profile showed her sodium to be 122 mmol/l, potassium 2.1 mmol/l, and urea 5.9 mmol/l. Serum creatinine was 182 μmol/l, phosphate 0.65 mmol/l, magnesium 0.59 mmol/l, and total corrected calcium 2.18 mmol/l. She was immediately given potassium and magnesium supplements, chlordiazepoxide, and intravenous vitamins including vitamin K and thiamine.

Three days after admission she developed a Staph aureus septicaemia secondary to a peripheral venous cannula infection. This required treatment with intravenous cefuroxime and flucloxacillin. She subsequently became drowsy and by day 10 had developed a severe spastic dysarthria and profound spastic tetraparesis. There was a bilateral lower motor neurone pattern of facial weakness and gaze evoked nystagmus. The clinical suspicion of CPM was supported by magnetic resonance imaging of the brain, which showed symmetrical signal hyperintensity in the pons on T2 weighted images, as well as generalised cerebral atrophy.

A review of the biochemistry results during her admission showed that the maximum increase in serum sodium concentration over a 24 hour period was only 7 mmol/l (from 123 to 130 mmol/l). Potassium and magnesium concentrations were corrected to the lower end of their normal ranges. However, she developed profound hypophosphataemia (0.16 mmol/l at nadir) which was rapidly corrected to 0.8 mmol/l within 72 hours. The rapid rise in plasma phosphate coincided with the onset of the patient’s neurological deterioration. With supportive care she made a gradual recovery such that two months after admission she was safe to be discharged, with only a mild residual left hemiparesis and slight spastic dysarthria, which were improving.

Case 2

A 44 year old woman was admitted with a three day history of progressive dysarthria, seven days of difficulty in walking, and dysaesthesia affecting all four limbs and the perioral region. She had consumed at least 80 units of alcohol a week for several months before presentation.

Examination on admission revealed a mild tetraparesis, dysarthria, and subjective sensory loss in both legs and the left arm. Her admission blood profile revealed a plasma sodium concentration of 136 mmol/l and potassium of 3.4 mmol/l. The serum phosphate concentration was profoundly low at 0.13 mmol/l. T2 weighted and FLAIR sequence MRI done three days after admission showed abnormal signal within the central brain stem suggestive of CPM (fig 1).

Figure 1

Coronal FLAIR magnetic resonance image (MRI) (A) and axial T2 weighted MRI (B) from case 2, showing high signal within the pons consistent with central pontine myelinosis.

She was treated with oral thiamine, multivitamins, and minerals including phosphate. She made a rapid improvement such that her dysarthria had resolved and gait improved sufficiently for her to be discharged 11 days after admission.


The pathophysiology of CPM is not well understood. Rapid correction of severe hyponatraemia is frequently implicated as a causative factor, but CPM has been reported in the presence of normonatraemia,1 hypokalemia,2 and hypophosphataemia.3 In these cases a hypothesis based on osmotic trauma must be questioned.

Recently an apoptotic hypothesis has been proposed.4 It is suggested that a depletion of the energy supply to glial cells might limit the function of their Na+/K+-ATPase pumps. This could reduce their ability to adapt to relatively minor osmotic stress caused by small changes in serum sodium concentration, and ultimately lead to apoptosis. A preliminary study of necropsy material from five cases of CPM compared with controls has provided some support for this theory. Using immunohistochemistry, an imbalance was shown between proapoptotic and antiapoptotic factors in glial cells with the appearance of oligodendrocytes.5 Furthermore the serum sodium concentrations in two of the patients remained normal from the onset of symptoms to the time of death.

The two patients presented here showed aclose temporal association between severe hypophosphataemia and the development of CPM. Both patients abused alcohol, and the first patient had moderate hyponatraemia with hypokalaemia. They may therefore have been particularly susceptible to CPM for a variety of reasons. It is possible, however, that severe hypophosphataemia adversely affected the Na+/K+-ATPase pump and finally triggered apoptosis and CPM. The temporal association of neurological deterioration with the rapid correction of profound hypophosphataemia in case 1 is unlikely to relate to osmotic stress in view of the small contribution of phosphate towards total osmolarity. The rapid change in plasma phosphate may, however, increase cellular stress, contributing to eventual apoptosis.

Both patients described here made good recoveries with phosphate replacement and supportive care. This suggests that widespread apoptosis had not occurred. In these patients the speed and degree of recovery might reflect the resolution of pontine oedema that could accompany less widespread or incomplete apoptosis.

There are useful practical conclusions to be drawn from the observed association of CPM with hypophosphataemia. First, one must suspect the diagnosis of CPM in susceptible patients even without “typical” electrolyte abnormalities. Second, as severe hypophosphataemia in itself has been correlated with increased mortality6 it would seem prudent to check and treat low serum phosphate concentrations in susceptible patients. This particularly refers to alcohol abusers or malnourished patients treated with intravenous glucose, diuretics, and steroids which may lower serum phosphate concentrations.


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.