Computer-aided diagnostic screen for Congenital Central Hypoventilation Syndrome with facial phenotype.

BACKGROUND: Congenital Central Hypoventilation Syndrome (CCHS) has devastating consequences if not diagnosed promptly. Despite identification of the disease-defining gene PHOX2B and a facial phenotype, CCHS remains underdiagnosed. This study aimed to incorporate automated techniques on facial photos to screen for CCHS in a diverse pediatric cohort to improve early case identification and assess a facial phenotype-PHOX2B genotype relationship. METHODS: Facial photos of children and young adults with CCHS were control-matched by age, sex, race/ethnicity. After validating landmarks, principal component analysis (PCA) was applied with logistic regression (LR) for feature attribution and machine learning models for subject classification and assessment by PHOX2B pathovariant. RESULTS: Gradient-based feature attribution confirmed a subtle facial phenotype and models were successful in classifying CCHS: neural network performed best (median sensitivity 90% (IQR 84%, 95%)) on 179 clinical photos (versus LR and XGBoost, both 85% (IQR 75-76%, 90%)). Outcomes were comparable stratified by PHOX2B genotype and with the addition of publicly available CCHS photos (n = 104) using PCA and LR (sensitivity 83-89% (IQR 67-76%, 92-100%). CONCLUSIONS: Utilizing facial features, findings suggest an automated, accessible classifier may be used to screen for CCHS in children with the phenotype and support providers to seek PHOX2B testing to improve the diagnostics. IMPACT: Facial landmarking and principal component analysis on a diverse pediatric and young adult cohort with PHOX2B pathovariants delineated a distinct, subtle CCHS facial phenotype. Automated, low-cost machine learning models can detect a CCHS facial phenotype with a high sensitivity in screening to ultimately refer for disease-defining PHOX2B testing, potentially addressing gaps in disease underdiagnosis and allow for critical, timely intervention.

Simple, miniaturized biosensors for wireless mapping of thermoregulatory responses.

Temperature is the most commonly collected vital sign in all of clinical medicine; it plays a critical role in care decisions related to topics ranging from infection to inflammation, sleep, and fertility. Most assessments of body temperature occur at isolated anatomical locations (e.g. axilla, rectum, temporal artery, or oral cavity). Even this relatively primitive mode for monitoring can be challenging with vulnerable patient populations due to physical encumbrances and artifacts associated with the sizes, weights, shapes and mechanical properties of the sensors and, for continuous monitoring, their hard-wired interfaces to data collection units. Here, we introduce a simple, miniaturized, lightweight sensor as a wireless alternative, designed to address demanding applications such as those related to the care of neonates in high ambient humidity environments with radiant heating found in incubators in intensive care units. Such devices can be deployed onto specific anatomical locations of premature infants for homeostatic assessments. The estimated core body temperature aligns, to within 0.05 °C, with clinical grade, wired sensors, consistent with regulatory medical device requirements. Time-synchronized, multi-device operation across multiple body locations supports continuous, full-body measurements of spatio-temporal variations in temperature and additional modes of determining tissue health status in the context of sepsis detection and various environmental exposures. In addition to thermal sensing, these same devices support measurements of a range of other essential vital signs derived from thermo-mechanical coupling to the skin, for applications ranging from neonatal and infant care to sleep medicine and even pulmonary medicine.

Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD): a collaborative review of the current understanding.

PURPOSE: To provide an overview of the discovery, presentation, and management of Rapid-onset Obesity with Hypothalamic dysfunction, Hypoventilation, and Autonomic Dysregulation (ROHHAD). To discuss a search for causative etiology spanning multiple disciplines and continents. METHODS: The literature (1965-2022) on the diagnosis, management, pathophysiology, and potential etiology of ROHHAD was methodically reviewed. The experience of several academic centers with expertise in ROHHAD is presented, along with a detailed discussion of scientific discovery in the search for a cause. RESULTS: ROHHAD is an ultra-rare syndrome with fewer than 200 known cases. Although variations occur, the acronym ROHHAD is intended to alert physicians to the usual sequence or unfolding of the phenotypic presentation, including the full phenotype. Nearly 60 years after its first description, more is known about the pathophysiology of ROHHAD, but the etiology remains enigmatic. The search for a genetic mutation common to patients with ROHHAD has not, to date, demonstrated a disease-defining gene. Similarly, a search for the autoimmune basis of ROHHAD has not resulted in a definitive answer. This review summarizes current knowledge and potential future directions. CONCLUSION: ROHHAD is a poorly understood, complex, and potentially devastating disorder. The search for its cause intertwines with the search for causes of obesity and autonomic dysregulation. The care for the patient with ROHHAD necessitates collaborative international efforts to advance our knowledge and, thereby, treatment, to decrease the disease burden and eventually to stop, and/or reverse the unfolding of the phenotype.

Ventilatory and Orthostatic Challenges Reveal Biomarkers for Neurocognition in Children and Young Adults With Congenital Central Hypoventilation Syndrome.

BACKGROUND: Children and young adults with congenital central hypoventilation syndrome (CCHS) are at risk of cognitive deficits. They experience autonomic dysfunction and chemoreceptor insensitivity measured during ventilatory and orthostatic challenges, but relationships between these features are undefined. RESEARCH QUESTION: Can a biomarker be identified from physiologic responses to ventilatory and orthostatic challenges that is related to neurocognitive outcomes in CCHS? STUDY DESIGN AND METHODS: This retrospective study included 25 children and young adults with CCHS tested over an inpatient stay. Relationships between physiologic measurements during hypercarbic and hypoxic ventilatory challenges, hypoxic ventilatory challenges, and orthostatic challenges and neurocognitive outcomes (by Wechsler intelligence indexes) were examined. Independent variable inclusion was determined by significant associations in Pearson's analyses. Multivariate linear regressions were used to assess relationships between measured physiologic responses to challenges and neurocognitive scores. RESULTS: Significant relationships were identified between areas of fluid intelligence and measures of oxygen saturation (SpO2) and heart rate (HR) during challenges. Specifically, perceptual reasoning was related to HR (adjusted regression [ß] coefficient, -0.68; 95% CI, 1.24 to -0.12; P = .02) during orthostasis. Working memory was related to change in HR (ß, -1.33; 95% CI, -2.61 to -0.05; P = .042) during the hypoxic ventilatory challenge. Processing speed was related to HR (ß, -1.19; 95% CI, -1.93 to -0.46; P = .003) during orthostasis, to baseline SpO2 (hypercarbic and hypoxic ß, 8.57 [95% CI, 1.63-15.51]; hypoxic ß, 8.37 [95% CI, 3.65-13.11]; P = .002 for both) during the ventilatory challenges, and to intrachallenge SpO2 (ß, 5.89; 95% CI, 0.71-11.07; P = .028) during the hypoxic ventilatory challenge. INTERPRETATION: In children and young adults with CCHS, SpO2 and HR-or change in HR-at rest and as a response to hypoxia and orthostasis are related to cognitive outcomes in domains of known risk, particularly fluid reasoning. These findings can guide additional research on the usefulness of these as biomarkers in understanding the impact of daily physical stressors on neurodevelopment in this high-risk group.

Transitional care and clinical management of adolescents, young adults, and suspected new adult patients with congenital central hypoventilation syndrome.

PURPOSE: With contemporaneous advances in congenital central hypoventilation syndrome (CCHS), recognition, confirmatory diagnostics with PHOX2B genetic testing, and conservative management to reduce the risk of early morbidity and mortality, the prevalence of identified adolescents and young adults with CCHS and later-onset (LO-) CCHS has increased. Accordingly, there is heightened awareness and need for transitional care of these patients from pediatric medicine into a multidisciplinary adult medical team. Hence, this review summarizes key clinical and management considerations for patients with CCHS and LO-CCHS and emphasizes topics of particular importance for this demographic. METHODS: We performed a systematic review of literature on diagnostics, pathophysiology, and clinical management in CCHS and LO-CCHS, and supplemented the review with anecdotal but extensive experiences from large academic pediatric centers with expertise in CCHS. RESULTS: We summarized our findings topically for an overview of the medical care in CCHS and LO-CCHS specifically applicable to adolescents and adults. Care topics include genetic and embryologic basis of the disease, clinical presentation, management, variability in autonomic nervous system dysfunction, and clarity regarding transitional care with unique considerations such as living independently, family planning, exposure to anesthesia, and alcohol and drug use. CONCLUSIONS: While a lack of experience and evidence exists in the care of adults with CCHS and LO-CCHS, a review of the relevant literature and expert consensus provides guidance for transitional care areas.

Neurocognition as a biomarker in the rare autonomic disorders of CCHS and ROHHAD.

PURPOSE: Congenital central hypoventilation syndrome (CCHS) and rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD) are rare disorders of autonomic regulation with risk for disrupted neurocognitive development. Our aim is to summarize research on neurocognitive outcomes in these conditions, advance understanding of how to best support these individuals throughout development, and facilitate future research. METHODS: We conducted a narrative review of literature on neurocognitive outcomes in CCHS and ROHHAD, supplemented with previously unpublished data from patients with CCHS and ROHHAD at our Center for Autonomic Medicine in Pediatrics (CAMP). RESULTS: Individuals with CCHS and ROHHAD experience a wide range of neurocognitive functioning ranging from above average to below average, but are at particular risk for difficulties with working memory, processing speed, perceptual reasoning, and visuographic skills. An assessment framework emphasizing fluid cognition seems especially appropriate for these conditions. Owing to small cohorts and varied methods of data collection, it has been difficult to identify associations between disease factors (including CCHS PHOX2B genotypes) and cognitive outcomes. However, results suggest that early childhood is a period of particular vulnerability, perhaps due to the disruptive impact of recurrent intermittent hypoxic episodes on brain and cognitive development. CONCLUSION: Neurocognitive monitoring is recommended as a component of routine clinical care in CCHS and ROHHAD as a marker of disease status and to ensure that educational support and disability accommodations are provided as early as possible. Collaborative efforts will be essential to obtain samples needed to enhance our understanding of neurocognitive outcomes in CCHS and ROHHAD.

Evolution of physiologic and autonomic phenotype in rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation over a decade from age at diagnosis.

Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD) is a rare cause of syndromic obesity with risk of cardiorespiratory arrest and neural crest tumor. No ROHHAD-specific genetic test exists at present. Rapid weight gain of 20-30 pounds, typically between ages 2-7 years in an otherwise healthy child, followed by multiple endocrine abnormalities herald the ROHHAD phenotype. Vigilant monitoring for asleep hypoventilation (and later awake) is mandatory as hypoventilation and altered control of breathing can emerge rapidly, necessitating artificial ventilation as life support. Recurrent hypoxemia may lead to cor pulmonale and/or right ventricular hypertrophy. Autonomic dysregulation is variably manifest. Here we describe the disease onset with "unfolding" of the phenotype in a child with ROHHAD, demonstrating the presentation complexity, need for a well-synchronized team approach, and optimized management that led to notable improvement ("refolding") in many aspects of the child's ROHHAD phenotype over 10 years of care. CITATION: Khaytin I, Stewart TM, Zelko FA, et al. Evolution of physiologic and autonomic phenotype in rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation over a decade from age at diagnosis. J Clin Sleep Med. 2022;18(3):937-944.

Machine learning mortality classification in clinical documentation with increased accuracy in visual-based analyses.

AIM: The role of machine learning on clinical documentation for predictive outcomes remains undefined. We aimed to compare three neural networks on inpatient providers' notes to predict mortality in neonatal hypoxic-ischaemic encephalopathy (HIE). METHODS: Using Children's Hospitals Neonatal Database, non-anomalous neonates with HIE treated with therapeutic hypothermia were identified at a single-centre. Data were linked with the initial seven days of documentation. Exposures were derived using the databases and applying convolutional and two recurrent neural networks. The primary outcome was mortality. The predictive accuracy and performance measures for models were determined. RESULTS: The cohort included 52 eligible infants. Most infants survived (n = 36, 69%) and 23 had severe HIE (44%). Neural networks performed above baseline and differed in their median accuracy for predicting mortality (P = .0001): recurrent models with long short-term memory 69% (25th , 75th percentile 65, 73%) and gated-recurrent model units 65% (62, 69%) and convolutional 72% (64, 96%). Convolutional networks' median specificity was 81% (72, 97%). CONCLUSION: The neural network models demonstrated fundamental validity in predicting mortality using inpatient provider documentation. Convolutional models had high specificity for (excluding) mortality in neonatal HIE. These findings provide a platform for future model training and ultimately tool development to assist clinicians in patient assessments and risk stratifications.