Objective: To compare the efficacy of a novel water-circulating surface cooling system with conventional measures for treating fever in neuro-intensive care unit patients.
Design: Prospective, unblinded, randomized controlled trial.
Setting: Neurologic intensive care unit in an urban teaching hospital.
Patients: Forty-seven patients, the majority of whom were mechanically ventilated and sedated, with fever > or =38.3 degrees C for >2 consecutive hours after receiving 650 mg of acetaminophen.
Interventions: Subjects were randomly assigned to 24 hrs of treatment with a conventional water-circulating cooling blanket placed over the patient (Cincinnati SubZero, Cincinnati OH) or the Arctic Sun Temperature Management System (Medivance, Louisville CO), which employs hydrogel-coated water-circulating energy transfer pads applied directly to the trunk and thighs.
Measurements and main results: Diagnoses included subarachnoid hemorrhage (60%), cerebral infarction (23%), intracerebral hemorrhage (11%), and traumatic brain injury (4%). The groups were matched in terms of baseline variables, although mean temperature was slightly higher at baseline in the Arctic Sun group (38.8 vs. 38.3 degrees C, p = .046). Compared with patients treated with the SubZero blanket (n = 24), Arctic Sun-treated patients (n = 23) experienced a 75% reduction in fever burden (median 4.1 vs. 16.1 C degrees -hrs, p = .001). Arctic Sun-treated patients also spent less percent time febrile (T > or =38.3 degrees C, 8% vs. 42%, p < .001), spent more percent time normothermic (T < or =37.2 degrees C, 59% vs. 3%, p < .001), and attained normothermia faster than the SubZero group median (2.4 vs. 8.9 hrs, p = .008). Shivering occurred more frequently in the Arctic Sun group (39% vs. 8%, p = .013).
Conclusion: The Arctic Sun Temperature Management System is superior to conventional cooling-blanket therapy for controlling fever in critically ill neurologic patients.