Machine learning classification of multiple sclerosis in children using optical coherence tomography.

BACKGROUND: In children, multiple sclerosis (MS) is the ultimate diagnosis in only 1/5 to 1/3 of cases after a first episode of central nervous system (CNS) demyelination. As the visual pathway is frequently affected in MS and other CNS demyelinating disorders (DDs), structural retinal imaging such as optical coherence tomography (OCT) can be used to differentiate MS. OBJECTIVE: This study aimed to investigate the utility of machine learning (ML) based on OCT features to identify distinct structural retinal features in children with DDs. METHODS: This study included 512 eyes from 187 (neyes = 374) children with demyelinating diseases and 69 (neyes = 138) controls. Input features of the analysis comprised of 24 auto-segmented OCT features. RESULTS: Random Forest classifier with recursive feature elimination yielded the highest predictive values and identified DDs with 75% and MS with 80% accuracy, while multiclass distinction between MS and monophasic DD was performed with 64% accuracy. A set of eight retinal features were identified as the most important features in this classification. CONCLUSION: This study demonstrates that ML based on OCT features can be used to support a diagnosis of MS in children.

Patterns of white and gray structural abnormality associated with paediatric demyelinating disorders.

The impact of multiple sclerosis (MS) and myelin oligodendrocyte glycoprotein (MOG) - associated disorders (MOGAD) on brain structure in youth remains poorly understood. Reductions in cortical mantle thickness on structural MRI and abnormal diffusion-based white matter metrics (e.g., diffusion tensor parameters) have been well documented in MS but not in MOGAD. Characterizing structural abnormalities found in children with these disorders can help clarify the differences and similarities in their impact on neuroanatomy. Importantly, while MS and MOGAD affect the entire CNS, the visual pathway is of particular interest in both groups, as most patients have evidence for clinical or subclinical involvement of the anterior visual pathway. Thus, the visual pathway is of key interest in analyses of structural abnormalities in these disorders and may distinguish MOGAD from MS patients. In this study we collected MRI data on 18 MS patients, 14 MOGAD patients and 26 age- and sex-matched typically developing children (TDC). Full-brain group differences in fixel diffusion measures (fibre-bundle populations) and cortical thickness measures were tested using age and sex as covariates. Visual pathway analysis was performed by extracting mean diffusion measures within lesion free optic radiations, cortical thickness within the visual cortex, and retinal nerve fibre layer (RNFL) and ganglion cell layer thickness measures from optical coherence tomography (OCT). Fixel based analysis (FBA) revealed MS patients have widespread abnormal white matter within the corticospinal tract, inferior longitudinal fasciculus, and optic radiations, while within MOGAD patients, non-lesional impact on white matter was found primarily in the right optic radiation. Cortical thickness measures were reduced predominately in the temporal and parietal lobes in MS patients and in frontal, cingulate and visual cortices in MOGAD patients. Additionally, our findings of associations between reduced RNFLT and axonal density in MOGAD and TORT in MS patients in the optic radiations imply widespread axonal and myelin damage in the visual pathway, respectively. Overall, our approach of combining FBA, cortical thickness and OCT measures has helped evaluate similarities and differences in brain structure in MS and MOGAD patients in comparison to TDC.

Physical Activity and Sedentary Behavior Patterns Across Weekdays and Weekend Days in Youth With Multiple Sclerosis and Controls.

Background: This study quantified and compared weekday and weekend patterns of device-measured physical activity (PA) and sedentary behavior between youth with pediatric multiple sclerosis (MS) and controls for the purpose of informing future PA behavior change interventions. Methods: Participant data were obtained from 3 ongoing observational studies, and the sample included 40 participants with pediatric MS and 41 controls. Light PA (LPA), moderate to vigorous PA (MVPA), and sedentary behavior data were collected using activity monitors (ActiGraph LLC) over 1 week. The main analysis involved a 2-way mixed factor analysis of variance with group as a between-subjects factor (pediatric MS vs control) and day as a within-subjects factor (weekday vs weekend day). Results: There was no group by day interaction from the analysis of variance for percentage of activity monitor wear time spent in LPA, MVPA, or sedentary behavior. There was no effect of group for LPA, MVPA, or sedentary behavior. There was an effect of day of week on percentage of day spent in LPA, MVPA, and sedentary behavior. Conclusions: These results suggest that youth with pediatric MS and controls were less physically active and more sedentary on weekends than on weekdays, but there were no differences between groups in PA and sedentary behavior overall or by day of the week. Physical activity interventions may be more successful by initially targeting weekend day activity.

When anorexia nervosa symptoms mask Kallmann syndrome.

BACKGROUND: Although anorexia nervosa might delay puberty, a structured assessment at its onset remains crucial in excluding congenital delayed puberty diseases. CASE PRESENTATION: During the follow-up of a 15-year-old girl suffering from anorexia nervosa, a change of treatment has led to a thorough medical history revealing the absence of the olfactory bulb. Kallmann syndrome diagnosis was made on a blood analysis and the patient was treated with a multidisciplinary approach. CONCLUSION: After the diagnosis, our patient was relieved as it has clarified some of her symptoms including anosmia, poor height and maturational delay. Too often a delayed puberty is attributed to anorexia nervosa itself without considering medical history. This case definitely shows the importance of performing a case history and early diagnosis in pre-pubertal AN to rule out other rare diseases and avoid mid- and long-term sequelae. LEVEL OF EVIDENCE: IV (case study).

Feasibility of a theory-informed mobile app for changing physical activity in youth with multiple sclerosis.

BACKGROUND: Youth with multiple sclerosis (MS) have high levels of disease activity, depression and fatigue, and lower moderate to vigorous physical activity (PA). PA participation is associated with lower disease activity, depression and fatigue as well as higher self-efficacy, goal setting and decreased barriers. These latter constructs may be targeted by an intervention program for behavior change, and such intervention must account for disease rarity, geographical proximity and time constraints which limit the feasibility, accessibility, and sustainability of implementing a PA intervention in youth with MS. We have developed a theory-informed mobile-app based PA promotion program to address these barriers, Active Teens with Multiple Sclerosis (ATOMIC) and herein report the feasibility of delivering this intervention in youth with MS. OBJECTIVE: To determine the feasibility of delivering the ATOMIC program to youth with MS. METHODS: Fifteen youth with MS (13F, 16.6 ±â€¯1.2 years) followed at the Hospital for Sick Children were enrolled in the ATOMIC intervention. Participants underwent a standard clinical evaluation; an exercise test to determine cardiorespiratory fitness; 7-day PA monitoring with an accelerometer; and completed standardized depression, fatigue questionnaires at baseline and post-intervention. Social cognitive scales related to self-efficacy, self-management, goal setting, perceived barriers, outcome expectancy, and social support were completed at baseline, six and 12-weeks. The 12-week mobile app PA intervention was informed by Social Cognitive Theory (SCT) and included: individualized PA coaching, PA self-monitoring (Fitbit), goal setting, social support and MS specific educational modules. Feasibility was defined as (1) meeting the recruitment target of 15 participants within a one-year period; (2) completion of ≥80% of study related questionnaires and testing; (3) a drop-out rate of less than 20%, and (4) adherence to the ATOMIC intervention program components of ≥80% (Fitbit wear, PA coaching calls, modules). RESULTS: From March 2018 to April 2019, 53% of youth approached agreed to participate (15/28). 13/15 participants completed the intervention. 36 of the possible 39 coaching calls (92%: 3 possible phone calls per participant); 89% wear adherence to FitbitCharge2 data (mean = 75 ±â€¯16.6 days of 84 days); and 5/12 (42%) of modules were completed. An average 8.4% (SD= 40%, range = -74.0 to 176%) increase in Fitbit steps above the first week of the intervention were observed. PA level from accelerometry at 12-weeks, aerobic fitness, depression and fatigue were unchanged. SCT scales indicated an increase in social support from friends (0.67-points, T = 2.7, pholm=0.04), and a decrease in outcome expectancy (-2.7, T = 3.0, pholm=0.03). There were no differences in self-efficacy, self-management or perceived barriers post intervention. CONCLUSIONS: Our results indicate that the ATOMIC program is feasible for delivery in youth with MS. Future work is needed to understand how to best implement each element of SCT through added features in the mobile-app, and to evaluate how the individual components of SCT mediate change in PA behaviours of youth with MS.

Pandemic-associated mental health changes in youth with neuroinflammatory disorders.

BACKGROUND: Children with neuroinflammatory disorders have high rates of anxiety and depression, alongside low rates of physical activity. Given general concerns for mental and physical health in children during the COVID-19 pandemic lockdown, here we sought to understand how sleep, anxiety, depression, and physical activity changed with the lockdown in children with neuroinflammatory disorders. We hypothesized that outcomes would worsen during the lockdown, and that they would differ by underlying disorder category and age. METHODS: Patients attending a specialized neuroinflammatory clinic (n = 314) completed questionnaires (n = 821 responses; Jan 2017-Aug 2020) assessing sleep, anxiety, depression, and physical activity. Respondents had either: childhood-onset chronic or recurrent neuroinflammatory disorders (CRNI), a history of Autoimmune Encephalitis (AE) or Monophasic Acquired Demyelinating Syndromes (monoADS). We performed linear mixed models to examine the association between our outcome measures (sleep, anxiety, depression, and physical activity) and categories of disorder type, sex, age, physical activity, relapses, and time (pre- vs. post- COVID-19 lockdown). Participant ID acted as a random effect, to account for repeated measures. RESULTS: Sleep significantly increased in the first 6 months of the COVID-19 lockdown (F(1, 544)=56.85, P<0.001,). Across the whole group, anxiety and depression did not change with the pandemic, but we found differing trends by age category. Anxiety decreased in teenagers (≥13y) (Z = 3.96, P<0.001), but not for pre-teens. Depression remained higher in teenagers than preteens across both timepoints (F(1, 597)=6.30, p = 0.012). Physical activity levels did not change with the pandemic in comparison to pre-pandemic (F(1, 629)=1.92, P = 0.166). Anxiety was higher in inactive individuals regardless of timing (F(2, 547)=3.74, p = 0.024). CONCLUSION: For youth with neuroinflammatory disorders, the COVID-19 pandemic lockdown resulted in increased hours of nighttime sleep but did not result in significant overall changes in self-reported anxiety or depression. Pre-lockdown, teenagers had higher depression and anxiety scores than preteens. Post-lockdown, anxiety and depression scores decreased in teenagers compared to pre-teens. Physical activity was low both pre- and post-lockdown, and rates of anxiety were higher for inactive participants at both timepoints. Differences based on age suggest that younger children (<13 years) were more negatively affected by the pandemic than older children (≥ 13 years).

Youth with multiple sclerosis have low levels of fitness.

BACKGROUND: Moderate and vigorous physical activity is associated with improved outcomes in youth with multiple sclerosis (MS). Physical fitness may also influence disease and health outcomes in this population. OBJECTIVES: To determine if there were differences in physical fitness between youth with MS and healthy controls (HC).To examine relationships between physical fitness, physical activity (PA) level, fatigue, depression and disease activity in youth with MS and HC. METHODS: Youth with MS (n = 19) and HC (n = 21) completed tests establishing cardiorespiratory-fitness (VO2peak), endurance via 2-minute walk test, and musculoskeletal strength via grip strength (GS). Questionnaires determined fatigue, depression, and PA levels. Weekly PA level was determined by accelerometry. Tests of differences and correlational analyses were used to evaluate physical fitness. RESULTS: Youth with MS had lower VO2peak (U = 279, p < 0.0001), endurance (t = 2.6, p = 0.02), and higher body mass index (BMI) (t = -5.9, p = 0.001) than HC. Higher VO2peak was associated with higher moderate to vigorous PAaccelerometer in HC (Spearman-Rho = 0.5, p = 0.03), but not in youth with MS (Spearman-Rho = 0.5, p = 0.06). Lower VO2peak and GS were associated with higher disability (Spearman-Rho = -0.6, p = 0.03) and relapses in MS (Spearman-Rho = -0.52, p = 0.04). CONCLUSIONS: Youth with MS have lower levels of fitness, compared with HC. Higher levels of fitness were associated with lower disease activity and disability in youth with MS.

Benefits of Physical Activity for Depression and Fatigue in Multiple Sclerosis: A Longitudinal Analysis.

OBJECTIVES: To examine the longitudinal relationship between physical activity and fatigue and depression among youth with demyelinating conditions. STUDY DESIGN: From September 2013 to March 2017, we performed a longitudinal study of consecutive youth diagnosed at their first visit with pediatric onset multiple sclerosis (POMS) or monophasic acquired demyelinating syndromes (mono-ADS) at a neuroinflammatory disorders clinic in a tertiary children's hospital. Fatigue was determined at each visit by the Pediatric Quality of Life Multidimensional Fatigue Scale, depressive symptoms by the Center of Epidemiologic Studies Depression Children Rating Scale, and physical activity level by the Godin Leisure Time Exercise Questionnaire. Mixed linear models were used to examine the associations of moderate-to-vigorous physical activity (MVPA) with fatigue and depression over time, adjusting for age, time from incident demyelination, sex, number of relapses, relapse within 30 days, and disability. RESULTS: In 182 patients (48 POMS, age 15 ± 1.7 years, 35 female; and 134 mono-ADS, age 12 ± 3.6 years 67 female) with 538 visits (mean follow-up 3.6 ± 2.7 years and 4.2 ± 3.3 years, respectively), a trajectory of increased fatigue over time was observed in POMS (2.28 points/year, P = .008) and mono-ADS (1.33 points/year, P = .007) patients. Youth with POMS had more depressive symptoms (estimate = 11.4 points, P < .002) than mono-ADS. Depressive symptoms increased over time in female patients with POMS (estimate = 1.4 points/year, P < .02). MVPA was associated with lower depression (-0.09, P < .001) and general fatigue (0.13, P = .02) over time in POMS. CONCLUSIONS: Youth with POMS who have higher levels of MVPA demonstrate lesser depressive symptoms and lower fatigue over time. Our results may inform future interventions to manage mood and fatigue in POMS.

Evaluation of an Ophthalmoscopy Simulator to Teach Funduscopy Skills to Pediatric Residents.

BACKGROUND: Medical school and residency training in ophthalmoscopic evaluation is limited, reducing diagnostic accuracy. We sought to evaluate the efficacy of self-study using an ophthalmoscopy simulator to improve the technical motor skills involved in direct funduscopy in postgraduate pediatric residents. METHODS: In this randomized-controlled study, 17 pediatric residents (postgraduate years 1-3) were randomized to control (n=8) or intervention (n=9) groups. Participants were asked to correctly identify the funduscopic findings presented to them on an ophthalmoscopy simulator after being trained on its use. Each participant was asked to review 20 images of the fundus, and then record their multiple-choice response on a scantron sheet listing all possible funduscopic pathologies. Pre- and post-intervention testing was performed. Survey data assessing exposure to funduscopy skills during undergraduate and postgraduate training and overall experience with the simulator were collected. RESULTS: Most (65% [11/17]) participants reported minimal or no formal teaching in ophthalmology during their undergraduate medical studies. Average pre-intervention score (of 20) was 10.24±1.75 (51%) for the entire group, with no statistically significant difference between average pre-score in the control (10.63±1.77) versus intervention (9.89±1.76, p=0.405) groups. Intervention subjects experienced a statistically significant improvement in scores (9.89±1.76 vs. 12.78±2.05, p=0.006 [95% confidence interval 4.80-0.98]), but control subjects did not. CONCLUSIONS: A single session with an ophthalmoscopy simulator can improve diagnostic accuracy in postgraduate pediatric trainees. Use of ophthalmoscopy simulation represents a novel addition to traditional learning methods for postgraduate pediatric residents that can help trainees to improve their confidence and accuracy in performing this challenging examination.