Marhl Marko, Grubelnik Vladimir, Magdič Marša, Markovič Rene
Faculty of Medicine, University of Maribor, SI-2000, Maribor, Slovenia; Faculty of Natural Sciences and Mathematics, University of Maribor, SI-2000, Maribor, Slovenia; Faculty of Education, University of Maribor, SI-2000, Maribor, Slovenia.
Faculty of Electrical Engineering and Computer Science, University of Maribor, SI-2000, Maribor, Slovenia.
Diabetes Metab Syndr. 2020 Jul-Aug;14(4):671-677. doi: 10.1016/j.dsx.2020.05.013. Epub 2020 May 8.
Clinical evidence exists that patients with diabetes are at higher risk for Coronavirus disease 2019 (COVID-19). We investigated the physiological origins of this clinical observation linking diabetes with severity and adverse outcome of COVID-19.
Publication mining was applied to reveal common physiological contexts in which diabetes and COVID-19 have been investigated simultaneously. Overall, we have acquired 1,121,078 publications from PubMed in the time span between 01-01-2000 and 17-04-2020, and extracted knowledge graphs interconnecting the topics related to diabetes and COVID-19.
The Data Mining revealed three pathophysiological pathways linking diabetes and COVID-19. The first pathway indicates a higher risk for COVID-19 because of a dysregulation of Angiotensin-converting enzyme 2. The other two important physiological links between diabetes and COVID-19 are liver dysfunction and chronic systemic inflammation. A deep network analysis has suggested clinical biomarkers predicting the higher risk: Hypertension, elevated serum Alanine aminotransferase, high Interleukin-6, and low Lymphocytes count.
The revealed biomarkers can be applied directly in clinical practice. For newly infected patients, the medical history needs to be checked for evidence of a long-term, chronic dysregulation of these biomarkers. In particular, patients with diabetes, but also those with prediabetic state, deserve special attention.
有临床证据表明，糖尿病患者感染2019冠状病毒病（COVID-19）的风险更高。我们研究了将糖尿病与COVID-19的严重程度和不良后果联系起来的这一临床观察结果的生理根源。
通过文献挖掘揭示糖尿病和COVID-19同时被研究的常见生理背景。总体而言，我们在2000年1月1日至2020年4月17日期间从PubMed获取了1,121,078篇文献，并提取了将与糖尿病和COVID-19相关的主题相互联系的知识图谱。
数据挖掘揭示了连接糖尿病和COVID-19的三条病理生理途径。第一条途径表明，由于血管紧张素转换酶2失调，感染COVID-19的风险更高。糖尿病和COVID-19之间的另外两个重要生理联系是肝功能障碍和慢性全身炎症。深入的网络分析提出了预测较高风险的临床生物标志物：高血压、血清丙氨酸转氨酶升高、白细胞介素-6升高和淋巴细胞计数降低。
所揭示的生物标志物可直接应用于临床实践。对于新感染的患者，需要检查病史以寻找这些生物标志物长期慢性失调的证据。特别是糖尿病患者以及糖尿病前期患者，值得特别关注。
