RESUMO
Gestational diabetes mellitus (GDM) poses increased risk of short- and long-term complications for mother and offspring1-4. GDM is typically diagnosed at 24-28 weeks of gestation, but earlier detection is desirable as this may prevent or considerably reduce the risk of adverse pregnancy outcomes5,6. Here we used a machine-learning approach to predict GDM on retrospective data of 588,622 pregnancies in Israel for which comprehensive electronic health records were available. Our models predict GDM with high accuracy even at pregnancy initiation (area under the receiver operating curve (auROC) = 0.85), substantially outperforming a baseline risk score (auROC = 0.68). We validated our results on both a future validation set and a geographical validation set from the most populated city in Israel, Jerusalem, thereby emulating real-world performance. Interrogating our model, we uncovered previously unreported risk factors, including results of previous pregnancy glucose challenge tests. Finally, we devised a simpler model based on just nine questions that a patient could answer, with only a modest reduction in accuracy (auROC = 0.80). Overall, our models may allow early-stage intervention in high-risk women, as well as a cost-effective screening approach that could avoid the need for glucose tolerance tests by identifying low-risk women. Future prospective studies and studies on additional populations are needed to assess the real-world clinical utility of the model.
Assuntos
Diabetes Gestacional/diagnóstico , Registros Eletrônicos de Saúde , Área Sob a Curva , Estudos de Coortes , Feminino , Humanos , Israel , Programas de Rastreamento , Gravidez , Prognóstico , Curva ROC , Reprodutibilidade dos Testes , Inquéritos e QuestionáriosRESUMO
Human gut microbiome composition is shaped by multiple factors but the relative contribution of host genetics remains elusive. Here we examine genotype and microbiome data from 1,046 healthy individuals with several distinct ancestral origins who share a relatively common environment, and demonstrate that the gut microbiome is not significantly associated with genetic ancestry, and that host genetics have a minor role in determining microbiome composition. We show that, by contrast, there are significant similarities in the compositions of the microbiomes of genetically unrelated individuals who share a household, and that over 20% of the inter-person microbiome variability is associated with factors related to diet, drugs and anthropometric measurements. We further demonstrate that microbiome data significantly improve the prediction accuracy for many human traits, such as glucose and obesity measures, compared to models that use only host genetic and environmental data. These results suggest that microbiome alterations aimed at improving clinical outcomes may be carried out across diverse genetic backgrounds.
Assuntos
Dieta/estatística & dados numéricos , Meio Ambiente , Características da Família , Microbioma Gastrointestinal/genética , Estilo de Vida , Adolescente , Adulto , Idoso , Idoso de 80 Anos ou mais , Estudos de Coortes , Feminino , Interação Gene-Ambiente , Glucose/metabolismo , Voluntários Saudáveis , Hereditariedade/genética , Humanos , Israel , Masculino , Pessoa de Meia-Idade , Obesidade/metabolismo , Fenótipo , Polimorfismo de Nucleotídeo Único/genética , RNA Bacteriano/análise , RNA Bacteriano/genética , RNA Ribossômico 16S/análise , Reprodutibilidade dos Testes , Estudos em Gêmeos como Assunto , Gêmeos/genética , Adulto JovemRESUMO
Aggregatibacter actinomycemcomitans, previously named Actinobacillus actinomycetemcomitans (Aa), is a facultative Gram-negative slow-growing coccobacillus associated with severe oral and nonoral infections. It is a member of the HACEK group. Pulmonary infection caused by Aa is rare. We describe two cases of Aa pneumonia mimicking malignancy and review published pediatric cases.