Panagiota Karatza
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Panagiota Karatza has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Dorottya Cserpan
Department of Neuropediatrics University Children's Hospital Zurich Zurich Switzerland
3 shared publications - Georgia Ramantani
Department of Neuropediatrics and Children's Research Center University Children's Hospital Zurich Zurich Switzerland
3 shared publications - Johannes Sarnthein
Zurich Neuroscience Center Eidgenössische Technische Hochschule Zürich and University of Zurich Zurich Switzerland
3 shared publications - Santo Pietro Lo Biundo
Department of Neuropediatrics and Children's Research Center University Children's Hospital Zurich Zurich Switzerland
3 shared publications - Francesco Pisani
Child Neurology and Psychiatry Unit, Department of Human Neurosciences Sapienza University of Rome Rome Italy
2 shared publications - Andrea Rüegger
Department of Neuropediatrics, University Children's Hospital Zurich, Zürich, Switzerland
1 shared publication
Latest Publications
- Open Access
Scalp high‐frequency oscillations differentiate neonates with hypoxic–ischemic encephalopathy with and without electroencephalographically confirmed seizures
Objective: Hypoxic–ischemic encephalopathy (HIE) is a leading cause of neonatal seizures, but seizure detection is challenging when continuous electroencephalography (EEG) is unavailable and neonates undergo only a single routine EEG. Biomarkers derived from routine EEG may aid identification of neonates with seizures. We investigated whether scalp high‐frequency oscillation (HFO) rates derived from the first eligible routine EEG (1) differentiate neonates with HIE with and without seizures; and (2) contribute to seizure classification alongside established clinical, neurophysiological, and imaging biomarkers. Methods: We retrospectively analyzed routine scalp EEG recordings from 33 term neonates with HIE (17 with seizures, 16 without), obtained at a mean postnatal age of 4 ± 3 days. The mean analyzed EEG duration was 38 ± 14 min in neonates with seizures and 36 ± 20 min in those without. HFO rates in sleep segments were quantified using a validated automated detector. Results: Scalp HFO rates were significantly higher in neonates with seizures than in those without (.12 ± .11 vs. .03 ± .02 HFOs/min/channel, adjusted p = .02). An HFO rate threshold of .07 HFOs/min/channel optimally discriminated neonates who had seizures from those who did not (area under the curve [AUC] = .81). HFO rates did not differ significantly according to HIE severity, therapeutic hypothermia, antiseizure medication response, EEG background activity, or magnetic resonance imaging (MRI) abnormalities. Multivariable models incorporating HFO rate, EEG background activity, and MRI findings showed discrimination (AUC = .92, 95% confidence interval [CI] = .69–1.00 for logistic regression; AUC = 1.00, 95% CI = .75–1.00 for random forest), with HFO rate as the strongest contributor to seizure classification. Significance: These findings support scalp HFO rate as a potential electrophysiological biomarker of seizure occurrence in neonatal HIE , complementary to EEG background assessment and MRI findings. Prospective studies are needed to determine whether HFOs can predict subsequent seizures.
Open AccessScalp high‐frequency oscillations differentiate neonates with hypoxic–ischemic encephalopathy with and without electroencephalographically confirmed seizures
Objective: Hypoxic–ischemic encephalopathy (HIE) is a leading cause of neonatal seizures, but seizure detection is challenging when continuous electroencephalography (EEG) is unavailable and neonates undergo only a single routine EEG. Biomarkers derived from routine EEG may aid identification of neonates with seizures. We investigated whether scalp high‐frequency oscillation (HFO) rates derived from the first eligible routine EEG (1) differentiate neonates with HIE with and without seizures; and (2) contribute to seizure classification alongside established clinical, neurophysiological, and imaging biomarkers. Methods: We retrospectively analyzed routine scalp EEG recordings from 33 term neonates with HIE (17 with seizures, 16 without), obtained at a mean postnatal age of 4 ± 3 days. The mean analyzed EEG duration was 38 ± 14 min in neonates with seizures and 36 ± 20 min in those without. HFO rates in sleep segments were quantified using a validated automated detector. Results: Scalp HFO rates were significantly higher in neonates with seizures than in those without (.12 ± .11 vs. .03 ± .02 HFOs/min/channel, adjusted p = .02). An HFO rate threshold of .07 HFOs/min/channel optimally discriminated neonates who had seizures from those who did not (area under the curve [AUC] = .81). HFO rates did not differ significantly according to HIE severity, therapeutic hypothermia, antiseizure medication response, EEG background activity, or magnetic resonance imaging (MRI) abnormalities. Multivariable models incorporating HFO rate, EEG background activity, and MRI findings showed discrimination (AUC = .92, 95% confidence interval [CI] = .69–1.00 for logistic regression; AUC = 1.00, 95% CI = .75–1.00 for random forest), with HFO rate as the strongest contributor to seizure classification. Significance: These findings support scalp HFO rate as a potential electrophysiological biomarker of seizure occurrence in neonatal HIE , complementary to EEG background assessment and MRI findings. Prospective studies are needed to determine whether HFOs can predict subsequent seizures.
- Open Access
Scalp high‐frequency activity differentiates neonates with seizures from healthy neonates and indicates postneonatal epilepsy risk
Objective: This study investigated whether scalp high‐frequency activity (HFA) rates in neonates with seizures predict postneonatal epilepsy (PNE). It also assessed whether HFA rates differentiate neonates with seizures from healthy neonates and whether they vary by seizure etiology, therapeutic hypothermia, and electroencephalographic (EEG) background activity. Methods: We included 47 neonates with EEG‐confirmed seizures (nine with neonatal mortality, three lost to follow‐up, 35 with 1‐year follow‐up), and eight healthy neonates. Scalp HFA rates during sleep were determined using an automated detector. Results: Neonatal seizure etiologies included hypoxic–ischemic encephalopathy (HIE, n = 16), structural vascular lesions (SVL, n = 14), and neonatal onset genetic epilepsies ( n = 14). Scalp HFA rates were significantly higher in neonates with seizures (.16 ± .15 HFA/min/channel [ch]) than in healthy neonates (.03 ± .02 HFA/min/ch), with a threshold of .06 HFA/min/ch best differentiating these groups. Among neonates with seizures, those with genetic etiologies had significantly higher HFA rates (.24 ± .19 HFA/min/ch) than those with SVL (.07 ± .05 HFA/min/ch). HFA rates were not associated with EEG background activity and were unaffected by therapeutic hypothermia in neonates with HIE. Of the 35 surviving neonates with seizures, 11 developed PNE, whereas 16 had normal development at follow‐up. Neonates who developed PNE had significantly higher HFA rates (.27 ± .18 HFA/min/ch) than those with normal development (.11 ± .09 HFA/min/ch), with a threshold of .12 HFA/min/ch best differentiating these groups. Significance: Scalp HFA differentiates neonates with seizures from healthy neonates and may help identify those at higher risk for PNE. These findings support the potential use of scalp HFA as a potential biomarker for seizure monitoring and epilepsy risk stratification in neonates.
Open AccessScalp high‐frequency activity differentiates neonates with seizures from healthy neonates and indicates postneonatal epilepsy risk
Objective: This study investigated whether scalp high‐frequency activity (HFA) rates in neonates with seizures predict postneonatal epilepsy (PNE). It also assessed whether HFA rates differentiate neonates with seizures from healthy neonates and whether they vary by seizure etiology, therapeutic hypothermia, and electroencephalographic (EEG) background activity. Methods: We included 47 neonates with EEG‐confirmed seizures (nine with neonatal mortality, three lost to follow‐up, 35 with 1‐year follow‐up), and eight healthy neonates. Scalp HFA rates during sleep were determined using an automated detector. Results: Neonatal seizure etiologies included hypoxic–ischemic encephalopathy (HIE, n = 16), structural vascular lesions (SVL, n = 14), and neonatal onset genetic epilepsies ( n = 14). Scalp HFA rates were significantly higher in neonates with seizures (.16 ± .15 HFA/min/channel [ch]) than in healthy neonates (.03 ± .02 HFA/min/ch), with a threshold of .06 HFA/min/ch best differentiating these groups. Among neonates with seizures, those with genetic etiologies had significantly higher HFA rates (.24 ± .19 HFA/min/ch) than those with SVL (.07 ± .05 HFA/min/ch). HFA rates were not associated with EEG background activity and were unaffected by therapeutic hypothermia in neonates with HIE. Of the 35 surviving neonates with seizures, 11 developed PNE, whereas 16 had normal development at follow‐up. Neonates who developed PNE had significantly higher HFA rates (.27 ± .18 HFA/min/ch) than those with normal development (.11 ± .09 HFA/min/ch), with a threshold of .12 HFA/min/ch best differentiating these groups. Significance: Scalp HFA differentiates neonates with seizures from healthy neonates and may help identify those at higher risk for PNE. These findings support the potential use of scalp HFA as a potential biomarker for seizure monitoring and epilepsy risk stratification in neonates.
- Open Access
Scalp high‐frequency oscillation spatial distribution is consistent over consecutive nights, while rates vary with antiseizure medication changes
Objective: This study aimed to investigate two key aspects of scalp high‐frequency oscillations (HFOs) in pediatric focal lesional epilepsy: (1) the stability of scalp HFO spatial distribution across consecutive nights, and (2) the variation in scalp HFO rates in response to changes in antiseizure medication (ASM).Methods: We analyzed 81 whole‐night scalp electroencephalography (EEG) recordings from 20 children with focal lesional epilepsy. We used a previously validated automated HFO detector to assess scalp HFO rates (80–250 Hz) during non–rapid eye movement (NREM) sleep. The spatial distribution of HFO rates across consecutive nights was evaluated using Hamming similarity, and changes in ASM were classified as increased, decreased, or stable.Results: For each patient, we analyzed 3 ± 1 whole‐night scalp EEG recordings, with a mean duration of 650 ± 215 min per recording. The distribution of HFO remained stable across consecutive nights, with a Hamming similarity of 88% ± 6%. Four patients had at least one ASM dosage decrease, nine patients had both ASM dosage decreases and increases, two patients had only ASM dosage increases, and five patients had no changes in ASM during the study period. A decrease in ASM dosage was associated with increased HFO rates (from .16 ± .32 to .22 ± .36 HFO/min; p = .03), whereas an increase in ASM dosage led to decreased HFO rates (from .32 ± .54 HFO/min to .22 ± .38 HFO/min; p = .005) when comparing the last night to the first.Significance: The spatial distribution of scalp HFOs remained consistent across multiple nights, whereas fluctuations in HFO rates correlated with changes in ASM dosage. These findings suggest that scalp HFOs may not only help identify epileptogenic brain tissue but also monitor treatment response.
Open AccessScalp high‐frequency oscillation spatial distribution is consistent over consecutive nights, while rates vary with antiseizure medication changes
Objective: This study aimed to investigate two key aspects of scalp high‐frequency oscillations (HFOs) in pediatric focal lesional epilepsy: (1) the stability of scalp HFO spatial distribution across consecutive nights, and (2) the variation in scalp HFO rates in response to changes in antiseizure medication (ASM).Methods: We analyzed 81 whole‐night scalp electroencephalography (EEG) recordings from 20 children with focal lesional epilepsy. We used a previously validated automated HFO detector to assess scalp HFO rates (80–250 Hz) during non–rapid eye movement (NREM) sleep. The spatial distribution of HFO rates across consecutive nights was evaluated using Hamming similarity, and changes in ASM were classified as increased, decreased, or stable.Results: For each patient, we analyzed 3 ± 1 whole‐night scalp EEG recordings, with a mean duration of 650 ± 215 min per recording. The distribution of HFO remained stable across consecutive nights, with a Hamming similarity of 88% ± 6%. Four patients had at least one ASM dosage decrease, nine patients had both ASM dosage decreases and increases, two patients had only ASM dosage increases, and five patients had no changes in ASM during the study period. A decrease in ASM dosage was associated with increased HFO rates (from .16 ± .32 to .22 ± .36 HFO/min; p = .03), whereas an increase in ASM dosage led to decreased HFO rates (from .32 ± .54 HFO/min to .22 ± .38 HFO/min; p = .005) when comparing the last night to the first.Significance: The spatial distribution of scalp HFOs remained consistent across multiple nights, whereas fluctuations in HFO rates correlated with changes in ASM dosage. These findings suggest that scalp HFOs may not only help identify epileptogenic brain tissue but also monitor treatment response.