Machine-learning derived algorithms for prediction of radiographic progression in early axial spondyloarthritis.

OBJECTIVES: To compare machine learning (ML) to traditional models to predict radiographic progression in patients with early axial spondyloarthritis (axSpA). METHODS: We carried out a prospective French multicentric DESIR cohort study with 5 years of follow-up that included patients with chronic back pain for <3 years, suggestive of axSpA. Radiographic progression was defined as progression at the spine (increase of at least 1 point of mSASSS scores/2 years) or at the sacroiliac joint (worsening of at least one grade of the mNY score between 2 visits). Statistical analyses were based on patients without any missing data regarding the outcome and variables of interest (295 patients).Traditional modelling: we performed a multivariate logistic regression model (M1); then variable selection with stepwise selection based on Akaike Information Criterion (stepAIC) method (M2), and Least Absolute Shrinkage and Selection Operator (LASSO) method (M3).ML modelling: using "SuperLearner" package on R, we modelled radiographic progression with stepAIC, LASSO, random forest, Discrete Bayesian Additive Regression Trees Samplers (DBARTS), Generalized Additive Models (GAM), multivariate adaptive polynomial spline regression (polymars), Recursive Partitioning And Regression Trees (RPART) and Super Learner. Accuracy of these models was compared based on their 10-fold cross-validated AUC (cv-AUC). RESULTS: 10-fold cv-AUC for traditional models were 0.79 and 0.78 for M2 and M3, respectively. The three best models in the ML algorithms were the GAM, the DBARTS and the Super Learner models, with 10-fold cv-AUC of: 0.77, 0.76 and 0.74, respectively. CONCLUSIONS: Two traditional models predicted radiographic progression as good as the eight ML models tested in this population.

MUC5B promoter variant rs35705950 and rheumatoid arthritis associated interstitial lung disease survival and progression.

BACKGROUND: The major risk factor for idiopathic pulmonary fibrosis (IPF), MUC5B rs35705950, was found to be associated with rheumatoid arthritis-associated interstitial lung disease (RA-ILD). Whilst the MUC5B rs35705950 T risk allele has been associated with better survival in IPF, its impact on RA-ILD prognosis remains to be determined. Our objective was to explore the influence of MUC5B rs35705950 on survival and progression in RA-ILD. METHODS: Through an international retrospective observational study, patients with RA-ILD were genotyped for the MUC5B rs35705950 variant and consecutive pulmonary function tests (PFTs) findings were collected. Longitudinal data up to a 10-year follow-up were considered and analyzed using mixed regression models. Proportional hazards and joint proportional hazards models were used to analyze the association of baseline and longitudinal variables with lung transplant-free survival. Significant progression of RA-ILD was defined as at least an absolute or relative 10% decline of forced vital capacity at 2 years from baseline. RESULTS: Out of 321 registered patients, 261 were included in the study: 139 women (53.3%), median age at RA-ILD diagnosis 65 years (interquartile range [IQR] 57 to 71), 151 ever smokers (59.2%). Median follow-up was 3.5 years (IQR 1.3 to 6.6). Mortality rate was 32% (95%CI 19 to 42) at 10 years. The MUC5B rs35705950 variant did not impact lung transplant-free survival (HR for the T risk allele carriers=1.26; 95%CI 0.61 to 2.62; P=0.53). Decline in pulmonary function at 2 years was not influenced by MUC5B rs35705950 (OR=0.95; 95%CI 0.44 to 2.05; P=0.89), irrespective of the HRCT pattern. CONCLUSION: In this study, the MUC5B rs35705950 promoter variant did not influence transplant- free survival or decline in pulmonary function in patients with RA-ILD.

Neuralgic amyotrophy and hepatitis E infection: 6 prospective case reports.

INTRODUCTION: Hepatitis E virus (HEV) represents the main cause of enterically transmitted hepatitis worldwide. It is known that neuralgic amyotrophy (NA) is one of the most frequent neurological manifestations of HEV. However, clinical, electrodiagnostic (EDX) and MRI characteristics, as well as long-term follow-up of HEV-related NA have not been fully described yet. CASE REPORTS: We describe longitudinally clinical, EDX, biological and MRI results of six cases of HEV-associated NA, diagnosed from 2012 to 2017. Patients were between the ages of 33 and 57 years old and had a positive HEV serology. Clinical patterns showed the whole spectrum of NA, varying from extensive multiple mononeuropathy damage to single mononeuropathy. EDX results showed that the patients totalised 26 inflammatory mononeuropathies (1 to 8 per patient). These involved classical nerves such as suprascapular (6/6 cases), long thoracic (5/6 cases) and accessory spinal nerves (2/6 cases) and, some less frequent more distal nerves like anterior interosseous nerve (3/6 cases), as well as some unusual ones such as the lateral antebrachial cutaneous nerve (1/6 case), sensory fibres of median nerve (1/6 case) and phrenic nerves (1/6 case). After 2 to 8 years, all nerves had clinically recovered (muscle examination above 3/5 on MRC scale for all muscles except in one patient). DISCUSSION: HEV should be systematically screened when NA is suspected, whatever the severity, if the onset is less than 4 months (before IgM HEV-antibodies disappear) and appears to be frequently associated with severe clinical and EDX pattern, without increasing the usual recovery time.

Voluntary winging of the scapula: Proposed diagnostic criteria.

We report a series of 10 patients with unilateral, dynamic, winged scapula (WS), without cause, that was diagnosed as voluntary winging of the scapula (VWS). We compared clinical, electrodiagnostic, and other examination data for 10 patients with VWS and 146 with dynamic WS-related neuromuscular disorders, to establish a detailed pattern of the VWS subtype. In VWS, electrodiagnostic and other examinations did not reveal any neuromuscular or orthopedic cause. Winging was dynamic, obvious, neither medial nor lateral, and mainly involved the inferior angle of the scapula, in young patients. VWS never appeared during floor push-ups. Patients could produce WS at will with the index and healthy shoulder, between 25° and 65° of anterior elevation, or with shoulder internal rotation. VWS is a benign disorder that can be distinguished from neuromuscular WS by normal electrodiagnostic results for muscles and nerves of both shoulders and two specific clinical tests.