An electroencephalography-based sleep index and supervised machine learning as a suitable tool for automated sleep classification in children.

STUDY OBJECTIVES: Although sleep is frequently disrupted in the pediatric intensive care unit, it is currently not possible to perform real-time sleep monitoring at the bedside. In this study, spectral band powers of electroencephalography data are used to derive a simple index for sleep classification. METHODS: Retrospective study at Erasmus MC Sophia Children's Hospital, using hospital-based polysomnography recordings obtained in non-critically ill children between 2017 and 2021. Six age categories were defined: 6-12 months, 1-3 years, 3-5 years, 5-9 years, 9-13 years, and 13-18 years. Candidate index measures were derived by calculating spectral band powers in different frequent frequency bands of smoothed electroencephalography. With the best performing index, sleep classification models were developed for two, three, and four states via decision tree and five-fold nested cross-validation. Model performance was assessed across age categories and electroencephalography channels. RESULTS: In total 90 patients with polysomnography were included, with a mean (standard deviation) recording length of 10.3 (1.1) hours. The best performance was obtained with the gamma to delta spectral power ratio of the F4-A1 and F3-A1 channels with smoothing. Balanced accuracy was 0.88, 0.74, and 0.57 for two-, three-, and four-state classification. Across age categories, balanced accuracy ranged between 0.83 and 0.92 and 0.72 and 0.77 for two- and three-state classification, respectively. CONCLUSIONS: We propose an interpretable and generalizable sleep index derived from single-channel electroencephalography for automated sleep monitoring at the bedside in non-critically ill children ages 6 months to 18 years, with good performance for two- and three-state classification. CITATION: van Twist E, Hiemstra FW, Cramer ABG, et al. An electroencephalography-based sleep index and supervised machine learning as a suitable tool for automated sleep classification in children. J Clin Sleep Med. 2024;20(3):389-397.

A Systematic Review of Neuromonitoring Modalities in Children Beyond Neonatal Period After Cardiac Arrest.

OBJECTIVES: Postresuscitation care in children focuses on preventing secondary neurologic injury and attempts to provide (precise) prognostication for both caregivers and the medical team. This systematic review provides an overview of neuromonitoring modalities and their potential role in neuroprognostication in postcardiac arrest children. DATA RESOURCES: Databases EMBASE, Web of Science, Cochrane, MEDLINE Ovid, Google Scholar, and PsycINFO Ovid were searched in February 2019. STUDY SELECTION: Enrollment of children after in- and out-of-hospital cardiac arrest between 1 month and 18 years and presence of a neuromonitoring method obtained within the first 2 weeks post cardiac arrest. Two reviewers independently selected appropriate studies based on the citations. DATA EXTRACTION: Data collected included study characteristics and methodologic quality, populations enrolled, neuromonitoring modalities, outcome, and limitations. Evidence tables per neuromonitoring method were constructed using a standardized data extraction form. Each included study was graded according to the Oxford Evidence-Based Medicine scoring system. DATA SYNTHESIS: Of 1,195 citations, 27 studies met the inclusion criteria. There were 16 retrospective studies, nine observational prospective studies, one observational exploratory study, and one pilot randomized controlled trial. Neuromonitoring methods included neurologic examination, routine electroencephalography and continuous electroencephalography, transcranial Doppler, MRI, head CT, plasma biomarkers, somatosensory evoked potentials, and brainstem auditory evoked potential. All evidence was graded 2B-2C. CONCLUSIONS: The appropriate application and precise interpretation of available modalities still need to be determined in relation to the individual patient. International collaboration in standardized data collection during the (acute) clinical course together with detailed long-term outcome measurements (including functional outcome, neuropsychologic assessment, and health-related quality of life) are the first steps toward more precise, patient-specific neuroprognostication after pediatric cardiac arrest.

MMN: from immunological cross-talk to conduction block.

Multifocal motor neuropathy (MMN) is a rare inflammatory neuropathy characterized by progressive, asymmetric distal limb weakness and conduction block (CB). Clinically MMN is a pure motor neuropathy, which as such can mimic motor neuron disease. GM1-specific IgM antibodies are present in the serum of approximately half of all MMN patients, and are thought to play a key role in the immune pathophysiology. Intravenous immunoglobulin (IVIg) treatment has been shown to be effective in MMN in five randomized placebo-controlled trials. Despite long-term treatment with intravenous immunoglobulin (IVIg), which is efficient in the majority of patients, slowly progressive axonal degeneration and subsequent muscle weakness cannot be fully prevented. In this review, we will discuss the current understanding of the immune pathogenesis underlying MMN and how this may cause CB, available treatment strategies and future therapeutic targets.

Pathophysiology of immune-mediated demyelinating neuropathies--Part II: Neurology.

In the second part of this review we deal with the clinical aspects of immune-mediated demyelinating neuropathies. We describe the relationship between pathophysiology and symptoms and discuss the pathophysiology of specific disease entities, including Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy, multifocal motor neuropathy, anti-myelin-associated glycoprotein neuropathy, and POEMS syndrome.

Pathophysiology of immune-mediated demyelinating neuropathies-part I: neuroscience.

This first article of this review deals with neuroscientific aspects of immune-mediated demyelinating neuropathies. It describes the anatomy and physiology of normal myelinated axons, methods of studying peripheral nerve physiology, pathophysiological consequences of demyelination or damage at the node of Ranvier, and the mechanisms that may lead to impaired axonal membrane dysfunction or axonal degeneration. This article (part I) will be followed by a second (part II) dealing with clinical aspects of these neuropathies.

Symptoms of activity-induced weakness in peripheral nervous system disorders.

Activity-induced weakness was reported in multifocal motor neuropathy (MMN) and chronic inflammatory demyelinating polyneuropathy (CIDP). This was attributed to activity-dependent conduction block (CB) arising in demyelinated axons. It is not known if activity-induced weakness is common, nor if it is specific for MMN and CIDP. We, therefore, carried out an investigation by questionnaire in 64 MMN patients, 52 CIDP patients, 48 progressive spinal muscular atrophy (PSMA) patients, and 30 normal subjects. Subjects were asked if they experienced an increase in weakness when performing 10 common tasks. The percentage of tasks causing activity-induced weakness was higher in the patient groups than in the normal subjects (p < 0.001). The risk of activity-induced weakness exceeding that in normal subjects was sixfold higher for each patient group when adjusted for sex, age, and a fatigue score. With further adjustment for scores of weakness and axon loss, no significant differences were found between the patient groups. In conclusion, activity-induced weakness is frequently reported in MMN and CIDP. It is, however, not specific for these neuropathies as PSMA patients reported it to the same extent.

Cold paresis in multifocal motor neuropathy.

Increased weakness during cold (cold paresis) was reported in single cases of multifocal motor neuropathy (MMN). This was unexpected because demyelination is a feature of MMN and symptoms of demyelination improve, rather than worsen, in cold. It was hypothesized that cold paresis in MMN does not reflect demyelination only, but may indicate the existence of inflammatory nerve lesions with permanently depolarized axons that only just conduct at normal temperature, but fail at lower temperatures. We investigated symptoms of cold paresis in 50 MMN patients, 48 chronic inflammatory demyelinating polyneuropathy (CIDP) patients, 35 progressive spinal muscular atrophy (PSMA) patients, and 25 chronic idiopathic axonal polyneuropathy patients. We also investigated symptoms of increased weakness during warmth (heat paresis). Cold paresis was reported more often than heat paresis. Cold paresis was most frequently reported in MMN. Multivariate analysis indicated that MMN patients had a 4- to 6-fold higher risk of reporting cold paresis than CIDP or PSMA patients. Because cold paresis is not consistent with demyelination, the lesions in MMN may involve other mechanisms than demyelination only. In conclusion, symptoms of cold paresis are common in peripheral nervous system disorders, particularly in MMN. This supports the above-described hypothesis.

Activity-dependent conduction block in chronic inflammatory demyelinating polyneuropathy.

Previous studies suggested that activity-dependent conduction block (CB) contributes to weakness in chronic inflammatory demyelinating polyneuropathy (CIDP). These studies, however, investigated only one nerve segment per patient, employed cervical magnetic stimulation which may be submaximal, included nerves with extremely low compound muscle action potentials (CMAPs) which precludes assessment of CB, and lacked predefined criteria for activity-dependent CB. Obtaining more robust evidence for activity-dependent CB is important because it may be treated pharmacologically. We investigated 22 nerve segments in each of 18 CIDP patients, employed supramaximal electrical stimulation, excluded nerves with markedly reduced CMAPs, and adopted criteria for activity-dependent CB. Each nerve was tested before and immediately after 60 s of maximal voluntary contraction (MVC) of the relevant muscle. Per nerve segment we calculated segmental area ratio: (area proximal CMAP)/(area distal CMAP). Per nerve we calculated total area ratio: (area CMAP evoked at Erb's point)/(area most distally evoked CMAP). MVC induced no change in mean area ratios and no activity-dependent CB according to our criteria, except for one segment. MVC induced increases in distal and proximal CMAP area and duration. In segments with demyelinative slowing, MVC induced an increase in CMAP duration prolongation. Thus, in CIDP, muscle activity induces virtually no CB, but it may increase temporal dispersion of nerve action potentials.

Activity-dependent conduction block in multifocal motor neuropathy.

Previous studies suggested that activity-dependent conduction block (CB) contributes to weakness in multifocal motor neuropathy (MMN). Obtaining more robust evidence for activity-dependent CB is important because it may be a novel target for treatment strategies. We performed nerve conduction studies in 22 nerve segments of 19 MMN patients, before and immediately after 60 seconds of maximal voluntary contraction (MVC) of the relevant muscle. We employed supramaximal electrical stimulation, excluded nerves with marked axonal loss, and adopted criteria for activity-dependent CB. Per segment, the segmental area ratio [area proximal compound muscle action potential (CMAP)/area distal CMAP] was calculated and, per nerve, total area ratio (area CMAP at Erb's point/area distal CMAP) was obtained. MVC induced no changes in mean area ratios and induced no activity-dependent CB. In segments with CB before MVC, the MVC induced increased duration prolongation. In MMN, MVC induced temporal dispersion but no activity-dependent CB.