Continuous hypertonic saline therapy and the occurrence of complications in neurocritically ill patients

Objective: To evaluate potential side effects of continuous hypertonic 3% saline (CHS) as maintenance fluid in patients with brain injury. Methods: Retrospective chart analysis of prospectively collected data. Patients. Patients admitted to the neurosurgical Intensive care unit for >4 days with traumatic brain injury, stroke, or subarachnoid hemorrhage with a Glasgow Coma Scale <9 and elevated intracranial pressure (ICP) or at risk of developing elevated ICP were included. Based on physician preference, one group was treated with 3% CHS at a rate of 1.5 mL/kg/bw as maintenance fluid. The other group received 0.9% normal saline (NS). Two percent saline was used in the CHS group to wean patients off 3% CHS or when sodium was above 155. Data on serum sodium, blood urea nitrogen, creatinine, ICP, infection rate, length of stay, rates of deep vein thrombosis, and pulmonary emboli and dural thrombosis were collected prospectively. Results: One hundred seven patients in the CHS group and 80 in the NS group met the inclusion criteria. The incidence of moderate hypernatremia (Na >155 mmol/L) and severe hypernatremia (Na >160 mmol/L) was significantly higher in the CHS therapy group than in the NS group. No significant relationship between CHS infusion and renal dysfunction was found. Moderate and severe hypernatremia was associated with a higher risk of elevated blood urea nitrogen and creatinine levels. Acute renal failure was not seen in these patients. A total of 53.3% in the CHS group and in 16.3% in the NS group (p < 0.0001) had raised ICP (>25 mm Hg), consistent with the physicians decision to use CHS in patients with elevated ICP. Conclusions. CHS therapy was not associated with an increased rate of infection, deep vein thrombosis, or renal failure. However, there was a significant risk of developing hypernatremia. We conclude that CHS administration in patients with severe injuries is safe as long as sodium levels are carefully monitored. (Crit Care Med 2009; 37:1433-1441)