ABSTRACT
The absence of a long COVID (LC) or post-acute sequelae of COVID-19 (PASC) diagnostic has profound implications for research and potential therapeutics given the lack of specificity with symptom-based identification of LC and the overlap of symptoms with other chronic inflammatory conditions. Here, we report a machine-learning approach to LC/PASC diagnosis on 347 individuals using cytokine hubs that are also capable of differentiating LC from chronic lyme disease (CLD). We derived decision tree, random forest, and gradient-boosting machine (GBM) classifiers and compared their diagnostic capabilities on a dataset partitioned into training (178 individuals) and evaluation (45 individuals) sets. The GBM model generated 89% sensitivity and 96% specificity for LC with no evidence of overfitting. We tested the GBM on an additional random dataset (106 LC/PASC and 18 Lyme), resulting in high sensitivity (97%) and specificity (90%) for LC. We constructed a Lyme Index confirmatory algorithm to discriminate LC and CLD.
Subject(s)
COVID-19 , Cytokines , Lyme Disease , Machine Learning , Humans , COVID-19/diagnosis , Lyme Disease/diagnosis , Diagnosis, Differential , Cytokines/metabolism , Chronic Disease , Sensitivity and Specificity , Male , Female , Algorithms , SARS-CoV-2/isolation & purification , Middle Aged , Post-Lyme Disease Syndrome/diagnosis , AdultABSTRACT
OBJECTIVE: This retrospective study examined whether changes in patient pre- and post-treatment symptoms correlated with changes in anti-neuronal autoantibody titers and the neuronal cell stimulation assay in the Cunningham Panel in patients with Pediatric Autoimmune Neuropsychiatric Disorder Associated with Streptococcal Infection (PANDAS), and Pediatric Acute-onset Neuropsychiatric Syndrome (PANS). METHODS: In an analysis of all tests consecutively performed in Moleculera Labs' clinical laboratory from April 22, 2013 to December 31, 2016, we identified 206 patients who were prescribed at least one panel prior to and following treatment, and who met the PANDAS/PANS diagnostic criteria. Patient follow-up was performed to collect symptoms and treatment or medical intervention. Of the 206 patients, 58 met the inclusion criteria of providing informed consent/assent and documented pre- and post-treatment symptoms. Clinician and parent-reported symptoms after treatment or medical intervention were categorized as "Improved/Resolved" (n = 34) or "Not-Improved/Worsened" (n = 24). These were analyzed for any association between changes in clinical status and changes in Cunningham panel test results. Clinical assay performance was also evaluated for reproducibility and reliability. RESULTS: Comparison of pre- and post-treatment status revealed that the Cunningham Panel results correlated with changes in patient's neuropsychiatric symptoms. Based upon the change in the number of positive tests, the overall accuracy was 86%, the sensitivity and specificity were 88% and 83% respectively, and the Area Under the Curve (AUC) was 93.4%. When evaluated by changes in autoantibody levels, we observed an overall accuracy of 90%, a sensitivity of 88%, a specificity of 92% and an AUC of 95.7%. Assay reproducibility for the calcium/calmodulin-dependent protein kinase II (CaMKII) revealed a correlation coefficient of 0.90 (p < 1.67 × 10-6) and the ELISA assays demonstrated test-retest reproducibility comparable with other ELISA assays. CONCLUSION: This study revealed a strong positive association between changes in neuropsychiatric symptoms and changes in the level of anti-neuronal antibodies and antibody-mediated CaMKII human neuronal cell activation. These results suggest there may be clinical utility in monitoring autoantibody levels and stimulatory activity against these five neuronal antigen targets as an aid in the diagnosis and treatment of infection-triggered autoimmune neuropsychiatric disorders. Future prospective studies should examine the feasibility of predicting antimicrobial and immunotherapy responses with the Cunningham Panel.
Subject(s)
Autoantibodies/blood , Autoimmune Diseases/blood , Autoimmune Diseases/diagnosis , Obsessive-Compulsive Disorder/blood , Obsessive-Compulsive Disorder/diagnosis , Streptococcal Infections/blood , Streptococcal Infections/diagnosis , Adolescent , Autoimmune Diseases/psychology , Child , Child, Preschool , Female , Follow-Up Studies , Humans , Male , Obsessive-Compulsive Disorder/psychology , Retrospective Studies , Streptococcal Infections/psychology , Young AdultABSTRACT
Sporadic late onset nemaline myopathy (SLONM) is a progressive myopathy of indeterminate etiology and poor outcome. If associated with a monoclonal gammopathy, SLONM carries a more unfavorable prognosis. Immunotherapy was unsuccessful. We report two HIV-negative SLONM/monoclonal gammopathy patients who improved following intravenous immunoglobulin (IVIg) treatment alone or in combination with immunosuppressant agents. This favorable response to treatment suggests that a dysimmune mechanism is operative in some SLONM individuals. We suggest that IVIg deserves initial consideration for SLONM therapy.
Subject(s)
Immunoglobulins, Intravenous/therapeutic use , Immunosuppressive Agents/therapeutic use , Myopathies, Nemaline/drug therapy , Drug Therapy, Combination , Humans , Male , Middle Aged , Myopathies, Nemaline/diagnosis , Prognosis , Treatment OutcomeSubject(s)
Cat-Scratch Disease/complications , Epilepsies, Partial/etiology , Adolescent , Cat-Scratch Disease/diagnosis , Cat-Scratch Disease/physiopathology , Electroencephalography , Epilepsies, Partial/diagnosis , Epilepsies, Partial/diagnostic imaging , Female , Humans , Magnetic Resonance Imaging , Radiopharmaceuticals , Technetium Tc 99m Exametazime , Tomography, Emission-Computed, Single-Photon , Tomography, X-Ray ComputedABSTRACT
The purpose of this study was to determine the effectiveness of the Charleston bending brace when compared with the thoracolumbosacral orthosis (TLSO or Boston) brace in the treatment of single-curve adolescent-type idiopathic scoliosis. The Charleston and TLSO braces were applied for approximately 8 nighttime hours and 18 to 22 hours per day, respectively. Treatment success was defined as improvement of curve deterioration with <5 degrees progression from the start of brace therapy until the conclusion of treatment, as well as the absence for the need to perform corrective surgery. The success rates were determined by Risser stage, initial angle, type of curvature, and sex of the patient. In addition, the success rate of the Charleston brace was assessed by analyzing the degree of initial correction. One hundred twenty-two patients (94 girls, 28 boys) were studied. Eighty-five patients were treated with the Charleston brace and 37 with the TLSO brace. Mean Cobb angle of curvature before bracing was 30.4 degrees. The curvature was lumbar in 60 patients, thoracic in 56, and thoracolumbar in 6. The average follow-up time was 23 months, with a minimum follow-up of 1 year. Surgery was performed in 11.8% and 13.5% of patients in the Charleston and TLSO groups, respectively. In this patient population, no significant difference in success rate was found between the groups.