Leslie Dervan, MD
Seattle Children's Hospital
Seattle, Washington
Disclosure information not submitted.
Alexander Doud, MD, MS
Clinical Assistant Professor, Neurology
University of Washington, United States
Disclosure information not submitted.
Jane Di Gennaro, MD, MS
Associate Professor, Pediatrics
University of Washington & Seattle Children's Hospital, United States
Disclosure information not submitted.
Giulia Benedetti, MD
Assistant Professor, Neurology
University of Washington & Seattle Children's Hospital, United States
Disclosure information not submitted.
Lindsey Morgan, MD
Assistant Professor, Neurology
University of Washington & Seattle Children's Hospital, United States
Disclosure information not submitted.
Catherine Amlie-Lefond, MD
Professor, Neurology
University of Washington & Seattle Children's Hospital, United States
Disclosure information not submitted.
R Watson, MD, MPH
Professor, Pediatrics
University of Washington & Seattle Children's Hospital, United States
Disclosure information not submitted.
Mark Wainwright, MD, PhD
Professor, Neurology
University of Washington & Seattle Children's Hospital, United States
Disclosure information not submitted.
Title: Heart rate variability and EEG characteristics in children with CAR T-associated neurotoxicity
Background: Neurotoxicity is a life-threatening complication of chimeric antigen receptor T-cell (CAR-T) therapies for cancer. We evaluated heart-rate variability (HRV) and quantitative electroencephalography (qEEG) characteristics in children admitted to the pediatric ICU who received CAR-T therapy and ≥24 hours of EEG monitoring. We hypothesized that reduced HRV and qEEG measures would be associated with severity of neurological symptoms and with circadian rhythm disruption.
Methods: We calculated the standard deviation of RR intervals (SDNN) and the root mean square of successive differences of RR intervals (RMSSD). We described spectral components of RR intervals, including low frequency power (LF), high frequency power (HF) and LF/HF power ratio (LFHFr). We used spectral components of the qEEG to describe the normalized alpha/delta band power ratio (ADr), which fluctuates in correlation with sleep and arousal states. Circadian rhythm was evaluated by comparing daytime to nighttime measurements. We used Student’s t-test to compare means, and Pearson correlation coefficient and linear regression with robust standard errors clustered by EEG to compare HRV to ADr.
Results: 6 patients had 20 individual 24-hour periods of EEG monitoring. Median patient age was 13 years (interquartile range (IQR) 9-18). Patients were admitted to the ICU a median of 6.5 (IQR 5-7) days following CAR-T therapy for cytokine release syndrome or new neurologic symptoms, including headache, tremor, encephalopathy, or seizure. The means of all HRV metrics were lower for children with severe encephalopathy (minimum GCS ≤8) vs. those without (GCS >8), including SDNN (22 vs. 58), RMSSD (30 vs. 81), LF (150 vs. 473), HF (152 vs. 615) and LFHFr (0.69 vs. 1.24), as was ADr (mean 3.1 vs. 8.1; all p< 0.005). ADr demonstrated circadian variability (mean 5.7 day vs. 4.9 night, p< 0.005), as did HRV metrics, though circadian variability in ADr did not differ by encephalopathy status. Among HRV metrics, LFHFr correlated best with ADr (r=0.18, p< 0.005) and LFHFr was associated with ADr in a linear regression model (coefficient=1.13, 95%CI 0.1 – 2.2, p=0.032).
Conclusions: HRV measures and ADr are associated with severity of encephalopathy in a high-risk patient cohort. HRV assessment may complement clinical assessment of patients with neurologic injury.