Submitted by Dr TAM Oi Yan on 6 April 2009
Department of Intensive Care, Pamela Youde Nethersole Eastern Hospital
 
hypernatremia and hyperglycemia

A 49 year old man with schizophrenia and type 2 DM was admitted for decreased general condition. He was dull looking and refused to talk according to her mother. He had chills, mild flu-like symptoms and repeated vomiting with decreased oral intake for few days before admission. He had no fever, no abdominal pain. His usual medications included Metformin and Diamicron.

On arrival to AED, he was profoundly dehydrated and emaciated with clouded consciousness. His GCS was E4V4M6. He was running high fever and had tachycardia. Otherwise, his BP was stable. He was neither dyspneic nor in respiratory distress and was saturated in room air. Physical examination of the respiratory system and the abdomen was essentially normal. CXR was clear. Hstix was high and urine was not immediately available for dipstix analysis. The clinical diagnosis was hyperglycemia crisis. He was taken over to ICU for further management.

Initial laboratory results were as follows: blood glucose 61mmol/L, Na 166 mmol/L, Ur 22.7 mmol/L, Cr 211umol/L (baseline normal), pH (ABG) 7.339, HCO3 25 mEq/L, serum osmolality 427, urine osmolality 619, urine ketone trace; which confirmed the diagnosis of hyperosmolar hyperglycemic state (HHS). The corrected Na was up to 182 mmol/L. Take estimated body weight of 60kg in this patient, the estimated free water deficit was 10.8L.

Free Water Deficit = Total Body Water x ((measured Na/Desired Na)-1) x weight in kg (TBW = 0.6 if male and 0.5 if female)

He was resuscitated with fluid replacement and insulin infusion which aimed at a gradual correction of the body fluid deficit, serum sodium and glucose level. Serum sodium and potassium level was monitored 4 hourly as any rapid correction of osmolality could result in cerebral edema.

Total 3.5L IV fluid was replaced to the patient in the initial 12 hours of admission with a gradual correction of the corrected sodium from 182mmol/L to 176mmol/L and blood glucose from 61mmol/L to 14mmol/L. In the subsequent 24 hours and 48 hours, further 3.5L and 3.8L fluid was replaced orally and by IV infusion. After rehydration, renal function of the patient returned to normal and his consciousness was much improved. He was cheerful and motivated in ward.

It has been suggested that, in cases of hyperglycaemic hyperosmolar syndrome, altered mental status is best predicted by serum sodium levels; while serum glucose levels alone are considered a poor indicator. See Figure for the trends of measured Na level, corrected Na level and glucose level with time.

Corrected Na = Na + 0.4 ([Glucose] - 5.5)

This is simplified adaptation of the Katz method (NEJM 1973; 289:843) which has a change in Na of 0.3 mmol/L per mmol change in glucose. Hypernatraemia can sometimes be disclosed only when the corrected sodium concentration for the dilutional effect of hyperglycaemia is taken into account.

A number of studies demonstrated that neurological symptoms may be absent in cases of severe gradually developing hyperglycaemia. This could be due to the capacity of the brain tissue to restore intracellular water by accumulating electrolytes. Furthermore, the brain cells are relatively permeable to glucose even in the absence of insulin. Therefore, hyperglycaemia per se does not produce severe hypertonicity in the CNS. By contrast, hypernatraemia itself causes severe cellular dehydration in the CNS. In addition, this state is associated with a rather slow compensatory accumulation of brain osmolar content. Thus, not only the serum osmolality level, but also the serum sodium concentration in particular, should be measured when evaluating a patient with extreme hyperglycaemia. A normal or elevated serum sodium value implies that substantial cellular dehydration has taken place and that the risk of coma or other neurologic abnormalities is high. In contrast, hyponatraemia suggests that cellular dehydration has not occurred or has occurred only to a limited extent.

The precipitation of this episode of hyperosmolar hyperglycemic state in this patient was most likely infective as he presented with some URTI symptoms. Another risk factor we suspected this patient might have was an impaired thirst sensation, such that he was unable to take fluids during dehydration.

 
Further reading:
1. Kitabchi AE, Umpierrez GE, Murphy MB, Kreisberg RA: Hyperglycemic crises in adult patients with diabetes: a consensus statement from the American Diabetes Association. Diabetes Care 29:2739-2748, 2006
2. Haralampos J Milionis, George L Liamis, Moses S Elisaf: Plasma sodium changes in the hyperglycaemic state: Clinical aspects of pathophysiology and management. Int J Diabetes & Metabolism (2001) 9:60- 66