Non-convulsive seizures in the encephalopathic critically ill cancer patient does not necessarily portend a poor prognosis

• Main Authors: Cristina Gutierrez, Merry Chen, Lei Feng, Sudhakar Tummala
• Format: Article
• Language: English
• Published: BMC 2019-12-01
• Series: Journal of Intensive Care
• Subjects: Encephalopathy; Cancer; Critically ill; Non-convulsive seizures; Status epilepticus; Mortality
• Online Access: https://doi.org/10.1186/s40560-019-0414-0

Abstract Background Non-convulsive status epilepticus (NCSE) is present in 10–30% of ICU patients with altered mental status (AMS) and is associated to poor outcomes. To our knowledge, there is no data describing the prevalence and outcomes of critically ill cancer patients with AMS associated to non-convulsive seizures (NCS) or NCSE. We aim to describe the outcomes and risk factors of critically ill cancer patients with encephalopathy associated with non-convulsive seizures (NCS). Methods This is a 3-year prospective observational study in a mixed oncological ICU at MD Anderson Cancer Center. Data of ICU patients with moderate to severe encephalopathy (Glasgow Coma Score < 13) that underwent EEG monitoring to rule out NCS were collected. Multivariate logistic regression was performed to identify risk factors and outcomes. Results Of the 317 patients with encephalopathy who underwent EEG monitoring, 14.5% had NCS. Known risk factors such as sepsis, CNS infection, antibiotics, and cardiac arrest were not associated with increased risk of NCS. Patients with NCS were more likely to have received recent chemotherapy (41.3% vs 21.4%; p = 0.0036), have a CNS disease (39% vs 24.4%; p = 0.035), and abnormal brain imaging (60.9% vs 44.6%; p = 0.041). Patients with lower SOFA scores, normal renal function, and absence of shock were likely to have NCS as the cause of their encephalopathy (p < 0.03). After multivariate analysis, only abnormal brain imaging and absence of renal failure were associated with NCS. Mortality was significantly lower in patients with non-convulsive seizures when compared to those without seizures (45.7% vs 64%; p = 0.022); however, there was no significant association of seizures and mortality on a multivariable logistic regression analysis. Conclusions NCS in critically ill cancer patients is associated with abnormalities on brain imaging and lower prevalence of organ failure. Diagnosis and treatment of NCS should be a priority in encephalopathic cancer patients, as they can have lower mortality than non-seizing patients. Opposite to other populations, NCS should not be considered a poor prognostic factor in critically ill encephalopathic cancer patients as they reflect a reversible cause for altered mentation.