Secondary hypothermia in patients with super-refractory status epilepticus managed with propofol and ketamine

Therapeutic hypothermia as a potent nonpharmacologic antiseizure therapy has been investigated experimentally in animal models and humans. Although induced hypothermia has been shown to be neuroprotective in acute convulsive status epilepticus, whether its use will translate into improved outcomes for patients with super-refractory nonconvulsive status epilepticus (SRNCSE) has been debated. No clinical data are available on the occurrence and prognostic impact of secondary hypothermia (s-HT) in patients with SRNCSE. With the possibility of core to periphery redistribution of heat with propofol and a centrally mediated dose-dependent fall in body temperature with ketamine, we aimed to investigate the incidence of s-HT events in patients with SRNCSE managed with propofol and ketamine and their impact on clinical outcomes.

We performed a retrospective observational analysis of consecutive patients with SRNCSE managed with propofol and/or ketamine in a single-center neurological intensive care unit between December 1, 2012 and December 31, 2015. Patients were divided according to the occurrence of hypothermia (temperature < 35.0 °C) into an s-HT group and a nonhypothermia (n-HT) group. Patients who received targeted temperature management therapy were excluded. We compared the demographics, comorbidities, treatment characteristics, and outcomes between groups.

Ninety-nine consecutive patients with SRNCSE managed with propofol and/or ketamine were identified during the study period. Twenty patients who received targeted temperature management were excluded, leaving a total of 79 patients for analysis. Hypothermia was observed in 52% (41/79) of the study population. Ketamine was used in 63/79 patients (80%). Ketamine infusion rates were higher and of longer duration among patients who developed s-HT compared with those who did not (mean dosage: 57.35 ± 26.6 mcg/kg/min vs 37.17 ± 15 mcg/kg/min, P = 0.001; duration: 116.36 ± 81.9 h vs 88 ± 89.7 h, P = 0.048). Propofol was used in 78/79 patients (99%), with no significant differences in characteristics between groups (mean dosage: 46.44 ± 20.2 mcg/kg/min vs 36.9 ± 12.9 mcg/kg/min, P = 0.058; duration: 125.43 ± 96.4 h vs 102.3 ± 87.1 h, P = 0.215). No significant differences in demographics, comorbidities, status epilepticus duration and resolution rates, and outcomes were observed between groups.

In this single-center retrospective analysis of patients whose SRNCSE is being treated, higher doses and longer durations of ketamine were associated with the occurrence of s-HT. Further investigation is warranted to clarify the thermogenic effects of ketamine and its effect on status epilepticus outcomes.