Pediatric status epilepticus: How common is cerebrospinal fluid pleocytosis in the absence of infection?

Enhancing Pressure Injury Surveillance Using Natural Language Processing

OBJECTIVE

This study assessed the feasibility of nursing handoff notes to identify underreported hospital-acquired pressure injury (HAPI) events.

METHODS

We have established a natural language processing-assisted manual review process and workflow for data extraction from a corpus of nursing notes across all medical inpatient and intensive care units in a tertiary care pediatric center. This system is trained by 2 domain experts. Our workflow started with keywords around HAPI and treatments, then regular expressions, distributive semantics, and finally a document classifier. We generated 3 models: a tri-gram classifier, binary logistic regression model using the regular expressions as predictors, and a random forest model using both models together. Our final output presented to the event screener was generated using a random forest model validated using derivation and validation sets.

RESULTS

Our initial corpus involved 70,981 notes during a 1-year period from 5484 unique admissions for 4220 patients. Our interrater human reviewer agreement on identifying HAPI was high ( κ = 0.67; 95% confidence interval [CI], 0.58-0.75). Our random forest model had 95% sensitivity (95% CI, 90.6%-99.3%), 71.2% specificity (95% CI, 65.1%-77.2%), and 78.7% accuracy (95% CI, 74.1%-83.2%). A total of 264 notes from 148 unique admissions (2.7% of all admissions) were identified describing likely HAPI. Sixty-one described new injuries, and 64 describe known yet possibly evolving injuries. Relative to the total patient population during our study period, HAPI incidence was 11.9 per 1000 discharges, and incidence rate was 1.2 per 1000 bed-days.

CONCLUSIONS

Natural language processing-based surveillance is proven to be feasible and high yield using nursing handoff notes.

Determining Normative Values for Cerebrospinal Fluid Profiles in Infants

BACKGROUND

Previous studies of reference values for cerebrospinal fluid (CSF) profiles have been limited by small sample size and few exclusion criteria.

OBJECTIVE

To determine age-specific normative CSF white blood cell count (WBC), glucose, and protein values in infants ≤90 days old.

METHODS

Performed a retrospective cross-sectional study of infants ≤90 days old who had a diagnostic lumbar puncture between 2008 and 2016. Infants with bacterial meningitis, bacteremia, UTI, positive CSF herpes simplex virus polymerase chain reaction (PCR) result, traumatic lumbar puncture, ventriculoperitoneal shunt, prematurity, recent seizure, previous antibiotic use, and history of a complex chronic condition were excluded for calculations to determine normative values. Data on demographics and CSF values (WBC with differential, protein, glucose, enterovirus PCR) were collected. CSF values were compared by age and by enterovirus PCR results using Kruskal-Wallis and Wilcoxon rank tests.

RESULTS

A total of 1029 out of 2000 patients were included and divided into 3 age groups: 0 to 28 days, 29 to 60 days, 61 to 90 days. CSF WBC values were significantly greater for 0- to 28-day old infants (median: 3, 95th percentile: 14) than for 29- to 60-day and 61- to 90-day old infants (median: 2 and 2; 95th percentile: 7 and 11, respectively) (P < .001). With each month of life, the median CSF protein significantly decreased and glucose significantly increased. In the CSF WBC differential, monocytes were found to be prevalent.

CONCLUSION

We determined age-specific normative components for CSF profile values for infants 0 to 90 days.

Cerebrospinal Fluid Pleocytosis Not Attributable to Status Epilepticus in First 24 Hours

BACKGROUND

Status epilepticus (SE) has traditionally been thought to cause cerebrospinal fluid (CSF) pleocytosis. However, attributing CSF pleocytosis solely to SE without addressing the underlying etiology may lead to poor outcomes. Leukocyte recruitment to CSF has been shown to peak around 24 hours after prolonged seizures in animal studies, suggesting that CSF pleocytosis within the first 24 hours of SE onset may be due to underlying causes. The goal of this study is to assess if SE is associated with CSF pleocytosis, independent of other causes within the first 24 hours of onset.

METHODS

We completed a historical cohort study of adult patients with SE admitted to the intensive care unit of Vancouver General Hospital between March 2010 and May 2019.

RESULTS

Of the 441 patients admitted with SE during the study period, 107 met our inclusion criteria leading to 111 lumbar punctures (LPs), with 4 patients receiving two LPs. CSF pleocytosis was seen in 12 of 72 patients who underwent an LP within the first 24 hours of SE onset. In all 12 patients, a secondary etiology for the pleocytosis was observed aside from SE. Of the six CSF samples collected after 24 hours of onset that demonstrated pleocytosis, four had no cause for pleocytosis other than SE.

CONCLUSIONS

In all 12 patients with CSF pleocytosis in the first 24 hours of onset of SE, an underlying etiology was identified. Therefore, any pleocytosis noticed within the first 24 hours of onset of refractory SE should not be attributed solely to SE.

Cerebrospinal fluid findings in non-infectious status epilepticus

OBJECTIVE

Ictal activity itself can cause pathological cerebrospinal fluid (CSF) findings. However, data regarding pathological CSF findings caused by SE itself to date remain scarce. We here evaluated the frequency and specificity of pathological CSF findings in non-infectious SE.

METHODS

We performed a retrospective analysis of CSF samples in adult patients with episodes of non-infectious SE, who had been admitted to the Department of Neurology, University Hospital of Cologne. The following parameters were assessed: cell count, protein, and lactate content, CSF/serum glucose quotient (Q_Glc), disturbances of blood-brain-barrier function assessed by CSF/serum albumin quotient (Q_Alb), and qualitative intrathecal IgG synthesis assessed by unmatched oligoclonal bands in CSF.

RESULTS

We analysed 54 episodes of non-infectious SE in which CSF had been obtained. CSF pleocytosis was infrequent (6%). Elevated CSF protein content was present in 44% of all cases, whereas elevated CSF lactate content was found in 23% of the cases. A decreased Q_Glc was present in 9%. Dysfunction of blood-brain-barrier (BBBD) was the most frequent pathological finding, amounting to 55%. Unmatched oligoclonal bands in CSF were seen in 10% of non-infectious SE. Further analysis revealed that elevated CSF protein content was found predominantly in recfractory SE (p = 0.04). Elevated CSF lactate content was associated with shorter latency between onset of SE and CSF retrieval (p = 0.004), positive history of epilepsy (p = 0.02) and an acute symptomatic etiology (p = 0.04). BBBD was also present more often in acute symptomatic SE (p = 0.001) and was the sole pathological CSF parameter associated with clinical outcome: presence of BBBD was associated with a less favorable outcome (p = 0.02).

SIGNIFICANCE

Non-infectious SE itself does not commonly cause CSF pleocytosis. Data suggest that the detection of CSF pleocytosis should prompt further diagnostics for an underlying infectious or neoplastic etiology. In contrast, elevation of CSF protein content and BBBD were found frequently in non-infectious SE.

Cerebrospinal Fluid Pleocytosis in Critical Care Patients With Seizures

OBJECTIVES

To assess the etiology of cerebrospinal fluid (CSF) pleocytosis in critical care patients with seizure(s) or status epilepticus (SE). Many previous studies, some performed decades ago, concluded that CSF pleocytosis may be entirely attributable to seizure activity.

METHODS

We undertook a retrospective chart review of adult patients with an admitting or acquired diagnosis of seizure(s) or SE in critical care units at the Winnipeg Health Sciences Centre between 2009 and 2012. Patients were identified through a critical care information database at a tertiary care center. We limited our study to patients who had lumbar punctures at our center within 5 days of seizure(s) or SE.

RESULTS

Of 426 patients with seizures in critical care units, 51 met the inclusion criteria. Seizure subtypes included focal seizures (5 or 10%), generalized seizures (14 or 27%), and SE (32 or 63%). Twelve (seven with SE) of the 51 (24%) were found to have CSF pleocytosis. A probable etiological cause for the CSF pleocytosis was identified in all 12 cases.

CONCLUSIONS

We conclude that seizures do not directly induce a CSF pleocytosis. Instead, the CSF pleocytosis more likely reflects the underlying acute or chronic brain process responsible for the seizure(s). This was not readily apparent in early studies without magnetic resonance imaging (MRI) of the brain and currently available laboratory investigations. An etiological cause of CSF pleocytosis must always be sought when patients present with seizures and it should never be assumed that seizures are the cause.

Effect of epileptic seizures on the cerebrospinal fluid--A systematic retrospective analysis

OBJECTIVE

Analyses of the cerebrospinal fluid (CSF) are obligatory when epileptic seizures manifest for the first time in order to exclude life-threatening causes or treatable diseases such as acute infections or autoimmune encephalitis. However, there are only few systematic investigations on the effect of seizures themselves on CSF parameters and the significance of these parameters in differential diagnosis.

METHODS

CSF samples of 309 patients with epileptic and 10 with psychogenic seizures were retrospectively analyzed. CSF samples were collected between 1999 and 2008. Cell counts, the albumin quotient, lactate and Tau-protein levels were determined. Findings were correlated with seizure types, seizure etiology (symptomatic, cryptogenic, occasional seizure), and seizure duration.

RESULTS

Pathological findings were only observed in patients with epileptic but not with psychogenic seizures. The lactate concentration was elevated in 14%, the albumin quotient in 34%, and the Tau protein level in 36% of CSF samples. Cell counts were only slightly elevated in 6% of patients. Different seizure types influenced all parameters except for the cell count: In status epilepticus highest, in simple partial seizures lowest values were seen. Symptomatic partial and generalized epileptic seizures had significantly higher Tau-protein levels than cryptogenic partial seizures. In patients with repetitive and occasional epileptic seizures, higher Tau-protein levels were seen than in those with psychogenic seizures. Duration of epileptic seizures was positively correlated with the albumin quotient, lactate and Tau-protein levels. High variability of investigated CSF parameters within each subgroup rendered a clear separation between epileptic and psychogenic seizures impossible.

SIGNIFICANCE

Elevated cell counts are infrequently observed in patients with epileptic seizures and should therefore not uncritically be interpreted as a postictal phenomenon. However, blood-CSF barrier disruption, increased glucose metabolism and elevation of neuronal damage markers are observed in considerable percentages of patients and depend on many factors such as etiology, seizure type and duration.