Paroxysmal sympathetic hyperactivity in the neurological intensive care unit.

OBJECTIVE

To assess the incidence and associations of paroxysmal sympathetic hyperactivity (PSH) among febrile patients in the neurological intensive care unit (NICU).

METHODS

Prospective data collection of consecutive patients admitted to an academic NICU for >48 hours. Fever was defined as body temperature >38.3 degrees C (101.0 degrees F) on at least one measurement for two consecutive days. PSH was defined by the transient presence of four of the following six criteria: fever, tachycardia, hypertension, tachypnea, excessive diaphoresis and extensor posturing or severe dystonia in the absence of other potential causes for these clinical signs.

Methods

Ninety-three patients were included. Traumatic brain injury (TBI) was the primary diagnosis in 43 patients (46%). PSH were diagnosed in 17 patients (18%). It occurred in 14/43 patients (33%) with TBI but only 3/50 patients (6%) with other primary disorders (p = 0.01). Patients with PSH were younger (mean age: 35 +/- 14 versus 51 +/- 18 years old in the rest of the population) (p = 0.01), although the difference did not reach statistical significance when only TBI patients were included in the analysis (mean age: 35 +/- 15 years old in TBI patients with PSH versus 46 +/- 19 years old in TBI patients without PSH; p = 0.07). Duration of fever in patients with PSH (mean: 10.5 +/- 7 days; median: 9 days) was significantly longer than in the rest of the population (mean: 5.1 +/- 3 days) (p<0.001); the difference remained significant when the analysis was restricted to TBI patients (p = 0.001).

CONCLUSION

Episodes of PSH occur in one-third of TBI patients but are rare in other acute neurological conditions. They are more frequent in younger patients and are associated with prolonged duration of fever.