Dataset Information

Using electronic health data to explore effectiveness of ICU EEG and anti-seizure treatment.

ABSTRACT:

Objectives

The purpose of this study was to examine critical care continuous electroencephalography (cEEG) utilization and downstream anti-seizure treatment patterns, their association with outcomes, and generate hypotheses for larger comparative effectiveness studies of cEEG-guided interventions.

Methods

Single-center retrospective study of critically ill patients (n = 14,523, age ≥18 years). Exposure defined as ≥24 h of cEEG and subsequent anti-seizure medication (ASM) escalation, with or without concomitant anesthetic. Exposure window was the first 7 days of admission. Primary outcome was in-hospital mortality. Multivariable analysis was performed using penalized logistic regression.

Results

One thousand and seventy-three patients underwent ≥24 h of cEEG within 7 days of admission. After adjusting for disease severity, ≥24 h of cEEG followed by ASM escalation in patients not on anesthetics (n = 239) was associated with lower in-hospital mortality (OR 0.76 [0.53-1.07]), though the finding did not reach significance. ASM escalation with concomitant anesthetic use (n = 484) showed higher odds for mortality (OR 1.41 [1.03-1.94]). In the seizures/status epilepticus subgroup, post cEEG ASM escalation without anesthetics showed lower odds for mortality (OR 0.43 [0.23-0.74]). Within the same subgroup, ASM escalation with concomitant anesthetic use showed higher odds for mortality (OR 1.34 [0.92-1.91]) though not significant.

Interpretation

Based on our findings we propose the following hypotheses for larger comparative effectiveness studies investigating the direct causal effect of cEEG-guided treatment on outcomes: (1) cEEG-guided ASM escalation may improve outcomes in critically ill patients with seizures; (2) cEEG-guided treatment with combination of ASMs and anesthetics may not improve outcomes in all critically ill patients.

SUBMITTER: Amerineni R

PROVIDER: S-EPMC8670316 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The purpose of this pilot study was to analyze treatment pathways of pediatric epilepsy using the common data model (CDM) based on electronic health record (EHR) data. We also aimed to reveal whether CDM analysis was feasible and applicable to epilepsy research. We analyzed the treatment pathways of pediatric epilepsy patients from our institute who underwent antiseizure medication (ASM) treatment for at least 2 years, using the Observational Medical Outcomes Partnership (OMOP)-CDM. Subgroup analysis was performed for generalized or focal epilepsy, varying age of epilepsy onset, and specific epilepsy syndromes. Changes in annual prescription patterns were also analyzed to reveal the different trends. We also calculated the proportion of drug-resistant epilepsy by applying the definition of seizure persistence after application of two ASMs for a sufficient period of time (more than 6 months). We identified 1,192 patients who underwent treatment for more than 2 years (mean ± standard deviation: 6.5 ± 3.2 years). In our pediatric epilepsy cohort, we identified 313 different treatment pathways. Drug resistance, calculated as the application of more than three ASMs during the first 2 years of treatment, was 23.8%. Treatment pathways and ASM resistance differed between subgroups of generalized vs. focal epilepsy, different onset age of epilepsy, and specific epilepsy syndromes. The frequency of ASM prescription was similar between onset groups of different ages; however, phenobarbital was frequently used in children with epilepsy onset < 4 years. Ninety-one of 344 cases of generalized epilepsy and 187 of 835 cases of focal epilepsy were classified as medically intractable epilepsy. The percentage of drug resistance was markedly different depending on the specific electro-clinical epilepsy syndrome [79.0% for Lennox-Gastaut syndrome (LGS), 7.1% for childhood absence epilepsy (CAE), and 9.0% for benign epilepsy with centrotemporal spikes (BECTS)]. We could visualize the annual trend and changes of ASM prescription for pediatric epilepsy in our institute from 2004 to 2017. We revealed that CDM analysis was feasible and applicable for epilepsy research. The strengths and limitations of CDM analysis should be carefully considered when planning the analysis, result extraction, and interpretation of results.

Project description:Background and objectiveEpilepsy is one of the most common neurological disorders, whose development is typically detected via early seizures. Electroencephalogram (EEG) is prevalently employed for seizure identification due to its routine and low expense collection. The stochastic nature of EEG makes manual seizure inspections laborsome, motivating automated seizure identification. The relevant literature focuses mostly on supervised machine learning. Despite their success, supervised methods require expert labels indicating seizure segments, which are difficult to obtain on clinically-acquired EEG. Thus, we aim to devise an unsupervised method for seizure identification on EEG.MethodsWe propose the first fully-unsupervised deep learning method for seizure identification on raw EEG, using a variational autoencoder (VAE). In doing so, we train the VAE on recordings without seizures. As training captures non-seizure activity, we identify seizures with respect to the reconstruction errors at inference time. Moreover, we extend the traditional VAE training loss to suppress EEG artifacts. Our method does not require ground-truth expert labels indicating seizure segments or manual feature extraction.ResultsWe implement our method using the PyTorch library and execute experiments on an NVIDIA V100 GPU. We evaluate our method on three benchmark EEG datasets: (i) intracranial recordings from the University of Pennsylvania and the Mayo Clinic, (ii) scalp recordings from the Temple University Hospital of Philadelphia, and (iii) scalp recordings from the Massachusetts Institute of Technology and the Boston Children's Hospital. To assess performance, we report accuracy, precision, recall, Area under the Receiver Operating Characteristics Curve (AUC), and p-value under the Welch t-test for distinguishing seizure vs. non-seizure EEG windows. Our approach can successfully distinguish seizures from non-seizure activity, with up to 0.83 AUC on intracranial recordings. Moreover, our algorithm has the potential for real-time inference, by processing at least 10 s of EEG in a second.ConclusionWe take the first successful steps in deep learning-based unsupervised seizure identification on raw EEG. Our approach has the potential of alleviating the burden on clinical experts regarding laborsome EEG inspections for seizures. Furthermore, aiding the identification of early seizures via our method could facilitate successful detection of epilepsy development and initiate antiepileptogenic therapies.

Project description:OBJECTIVES:Shared decision-making is recommended for critically ill adults who face major, preference-sensitive treatment decisions. Yet, little is known about when and how patients and families are engaged in treatment decision-making over the longitudinal course of a critical illness. We sought to characterize patterns of treatment decision-making by evaluating clinician discourse in the electronic health record of critically ill adults who develop chronic critical illness or die in an ICU. DESIGN, SETTING, AND PATIENTS:We conducted qualitative content analysis of the electronic health record of 52 adult patients, admitted to a medical ICU in a tertiary medical center from January 1, 2016, to December 31, 2016. We included patients who met a consensus definition of chronic critical illness (26 patients) and a matched sample who died or transitioned to hospice care in the ICU before developing chronic critical illness (26 patients). INTERVENTIONS:None. MEASUREMENTS AND MAIN RESULTS:Characterization of clinician decision-making discourse documented during the course of an ICU stay. Clinician decision-making discourse in the electronic health record followed a single, consistent pattern across both groups. Initial decisions about admission to the ICU focused on specific interventions that can only be provided in an ICU environment (intervention-focused decisions). Following admission, the documented rationale for additional treatments was guided by physiologic abnormalities (physiology-centered decisions). Clinician discourse transitioned to documented engagement of patients and families in decision-making when treatments failed to achieve specified physiologic goals. The phrase "goals of care" is common in the electronic health record and is used to indicate poor prognosis, to describe conflict with families, and to provide rationale for treatment limitations. CONCLUSIONS:Clinician discourse in the electronic health record reveals that patient physiology strongly guides treatment decision-making throughout the longitudinal course of critical illness. Documentation of patient and family engagement in treatment decision-making is limited until available medical treatments fail to achieve physiologic goals.

Project description:ObjectivesWe tested the hypothesis that routine monitoring data could describe a detailed and distinct pathophysiologic phenotype of impending hypoglycemia in adult ICU patients.DesignRetrospective analysis leading to model development and validation.SettingAll ICU admissions wherein patients received insulin therapy during a 4-year period at the University of Virginia Medical Center. Each ICU was equipped with continuous physiologic monitoring systems whose signals were archived in an electronic data warehouse along with the entire medical record.PatientsEleven thousand eight hundred forty-seven ICU patient admissions.InterventionsThe primary outcome was hypoglycemia, defined as any episode of blood glucose less than 70 mg/dL where 50% dextrose injection was administered within 1 hour. We used 61 physiologic markers (including vital signs, laboratory values, demographics, and continuous cardiorespiratory monitoring variables) to inform the model.Measurements and main resultsOur dataset consisted of 11,847 ICU patient admissions, 721 (6.1%) of which had one or more hypoglycemic episodes. Multivariable logistic regression analysis revealed a pathophysiologic signature of 41 independent variables that best characterized ICU hypoglycemia. The final model had a cross-validated area under the receiver operating characteristic curve of 0.83 (95% CI, 0.78-0.87) for prediction of impending ICU hypoglycemia. We externally validated the model in the Medical Information Mart for Intensive Care III critical care dataset, where it also demonstrated good performance with an area under the receiver operating characteristic curve of 0.79 (95% CI, 0.77-0.81).ConclusionsWe used data from a large number of critically ill inpatients to develop and externally validate a predictive model of impending ICU hypoglycemia. Future steps include incorporating this model into a clinical decision support system and testing its effects in a multicenter randomized controlled clinical trial.