Managing hyperosmolar hyperglycaemic syndrome

Abstract

I would like to comment on the Association of British Clinical Diabetologists (ABCD) guidelines for the management of hyperglycaemic emergencies in adults (Practical Diabetes International June 2006, pages 227-231). Although the guidelines are simply and straightforwardly presented, I feel they seriously underplay the severity of hyperosmolar nonketotic coma (which incidentally is a misleading term: it should be abandoned and replaced by hyperosmolar hyperglycaemic syndrome, HHS). I think it is a mistake to describe the management ‘as for DKA [diabetic ketoacidosis] but …’ as there are some significant differences which need to be emphasised. Failure to understand the underlying disorder may lead to inappropriate management. There is no precise or internationally agreed definition of HHS but there is recognition that minor degrees of ketosis and acidosis (arterial pH >7.3, bicarbonate >15mmol/L) are common. Patients with HHS are often elderly and the condition may be the first presentation of type 2 diabetes. It appears to be more common in people of African or Afro-Caribbean origin. In recent years, there have been case reports of HHS occurring in children and young adults — not all with type 2 diabetes. Severe dehydration and pre-renal failure are usual. Unlike DKA, the onset is over many days and, although it is correct to say that vascular complications are common, cerebral oedema is the most serious complication and may present as seizures, progressive confusion, or altered consciousness and lead to permanent neurological damage or death. Correct interpretation of the serum Na is important. Total body Na is low and there may be marked water deficiency. Although the initial Na may be normal or high, it is important to understand that high concentrations of effective solutes such as glucose will cause fluid shifts out of the cell and a dilutional fall in Na. The corrected Na can be calculated by adding 2.4mmol/L for every 5.5mmol/L where the glucose is above normal but, more importantly, it follows that as treatment results in a fall in glucose, there will almost inevitably be a rise in serum Na. Therefore, the rise in Na is expected and predictable — what is important is what is happening to tonicity. Thus, a rising Na alone is not an indication for the administration of hypotonic fluids. Care must be taken to avoid situations which may increase the risk of cerebral oedema. Extracellular hypertonicity causes cellular dehydration which in the brain is resisted by a protective mechanism known as cerebral volume regulation. Water loss is limited by the production of intracellular osmotically active ions and, later, organic osmolytes or idiogenic osmoles. Whilst this rise in cellular osmolality is beneficial in the hyperosmolar state, rapid rehydration with IV fluids may lead to water movement into the cell and swelling occurs before compensatory mechanisms have time to eliminate these solutes. This may be the mechanism which leads to cerebral oedema and is the reason why many argue that hypotonic solutions should be avoided (there is no experimental evidence to justify using less than 0.5N saline) and that slower correction of the hyperosmolality is safer than rapid correction. Assessing the level of dehydration is difficult and often over-estimated. Postural changes in heart rate and BP can be helpful but not always practical. It is rare to need more than twice the usual daily fluid requirements. The aim is to replace approximately 50% of the deficit in the first 12 hours. These patients are often transferred too soon to general medical wards and fluid management left to inexperienced junior medical staff. Severe HHS should be seen soon after admission by diabetes specialists with experience of managing this challenging emergency, and close supervision given until the patient is eating and drinking normally. Adrian Scott*, * Diabetes Centre, Northern General Hospital, Herries Road, Sheffield S5 7AU, UK