Clinical data on the types of respiratory pathogens which are most frequently engaged in respiratory co-infections of children and adults are lacking. We analyzed 10 years of data on a total of over 15,000 tests for 16 viral and bacterial pathogens detected in clinical samples at the University Hospital of Augsburg, Germany. Co-infection frequencies and their seasonal patterns were examined using a proportional distribution model. Co-infections were detected in 7.3% of samples, with a higher incidence in children and males. The incidence of interbacterial and interviral co-infections was higher than expected, whereas bacterial-viral co-infections were less frequent. H. influenzae, S. pneumoniae, rhinovirus, and respiratory syncytial virus (RSV) were most frequently involved. Most co-infections occurred in winter, but distinct summer peaks were also observed, which occurred even in children, albeit less pronounced than in adults. Seasonality of respiratory (co-)infections decreased with age. Our results suggest to adjust existing testing strategies during high-incidence periods.

Background/Objectives: The role of maternal diet in atopic dermatitis (AD) requires better understanding, as AD often manifests early in life and precedes other allergic diseases. We evaluated the association between maternal diet and AD up to 2 years of age. Methods: A total of 116 mother–child dyads from the CARE birth cohort study were included. Maternal diet during pregnancy was assessed with a validated self-administered 97-item food frequency questionnaire, and dietary scores were calculated. AD was evaluated at ages 4 months, 1 year, and 2 years. The associations between maternal dietary patterns and AD were examined by logistic regression analysis adjusting for total energy intake, gender of the child, maternal antibiotic therapy during pregnancy, and history of atopic disease among both parents. Results: Of the 116 children, 27 (23.3%) developed AD by 2 years, 11 of whom (40.7%) had persistent AD within the first 2 years. AD risk was reduced with a higher Mediterranean diet score during pregnancy (upper median > 3 points versus lower median: adjusted OR 0.24, 95% CI 0.08–0.69, p = 0.009) and with greater dietary diversity, as measured by the number of items consumed (upper median > 53 items versus lower median: OR 0.19, 95% CI 0.06–0.58, p = 0.005). No association was found with macronutrients and micronutrients. Red meat consumption showed a positive association with the persistent AD phenotype (adjusted OR 5.04, 95% CI 1.47 to 31.36, p = 0.034). Conclusions: Adherence to a Mediterranean diet and a diverse diet during pregnancy may decrease the risk of developing early childhood AD. This highlights the synergistic role of nutrients in dietary patterns as they modulate immune development and disease susceptibility.

Background: A critical lesson from the public health crisis caused by the COVID-19 pandemic is that pandemic preparedness requires rapid identification of early biomarkers to predict need of hospitalization. Furthermore, while therapy for COVID-19 is available, it must be administered early after infection and targeted specifically to patients in need. Currently, patient demographics and pre-existing clinical conditions are used to determine treatment eligibility. However, these factors do not provide accurate predictions, and previous research on COVID-19 severity biomarkers has focused on hospitalized patients. Here, we identify point-of-care biomarkers capable of accurately predicting COVID-19-related hospitalization risk and disease progression shortly after a positive diagnosis.Methods: Previously uninfected and non-vaccinated individuals (N=201) were recruited at, or shortly after, SARS-CoV-2 diagnosis. Biochemistry and serum cytokine levels were longitudinally measured. Biomarker combinations were tested by multi-dimensional clustering followed by dimension reduction.Findings: The combination of high levels of LDH or serum IL-6 showed the most accurate prediction (sensitivity=100%, specificity=94.5%) for hospitalization risk. The combination of four cytokines (IFN-a, IFN-g, IL-6, IL-8) was the most accurate predictor for moderate versus mild disease. Age and BMI showed significantly lower predictive values for hospitalization risk and were not predictive of mild versus moderate symptoms.Interpretation: Early accurate prediction of COVID-19 hospitalization and severity is possible using biomarkers feasibly measurable at point-of-care, thus allowing optimizing personalized treatment and public health management. The methodology used here sets out a paradigm for severity prediction in the framework of pandemic preparedness.