估计流感后症状、病症和疾病的范围与持续时间。

文献信息

中文摘要

背景： 流感的急性症状已为人们所了解；然而，对于潜在的急性后遗症知之甚少。我们进行了一项回顾性匹配队列研究，以识别流感后风险升高的病症。
方法： 我们分析了Merative MarketScan研究数据库（2001-2019）的数据，包括330万被诊断为流感的患者。病例按相似年龄、性别、诊疗环境和入组时期与无流感诊断的患者进行1:1匹配。我们检查了索引流感事件前后一年内诊断编码的频率，应用K-means聚类来识别与咳嗽（一种已知的流感后遗症）具有相似流感后就诊模式的诊断。对病症进行审查并汇总为临床类别。我们使用双重差分法比较了流感病例与对照之间的诊断趋势。
结果： 我们识别出67个在流感感染后发病率显著增加的诊断类别。既包括众所周知的后续病症，如肺炎和中耳炎，也包括较少预期的发现，如急性肾衰竭和口腔念珠菌病。风险升高的持续时间因病症而异，从一周（例如横纹肌溶解症）到5周（例如咯血）或8周（例如慢性支气管炎）不等。
结论： 我们发现流感与通常不被认为相关的病症有关。此外，一些症状的持续时间比通常认为的更长，提示部分患者可能因流感感染而遭受迁延性疾病。我们的发现不仅对展示流感的全部负担具有重要意义，也为未来预防或治疗流感的研究提供了信息。

英文摘要

BACKGROUND: Acute symptoms of influenza are understood; however, less is known about potential postacute sequelae. We conducted a retrospective matched cohort study to identify conditions with elevated risk following influenza.
METHODS: We analyzed data from the Merative MarketScan Research Databases (2001-2019), including 3.3 million patients with a diagnosis of influenza. Cases were matched 1:1 to patients based on similar age, sex, care setting, and enrollment period with no influenza diagnosis. We examined the frequency of diagnosis codes in the year before and after the index influenza event, applying K-means clustering to identify diagnoses with post-influenza visit patterns similar to cough, a known sequela of influenza. Conditions were reviewed and aggregated into clinical categories. We compared trends in diagnoses between influenza cases and controls using a difference-in-differences approach.
RESULTS: We identified 67 diagnostic categories with significantly increased incidence post-influenza infection. Both commonly known sequelae, such as pneumonia and otitis media as well as less expected findings, such as acute kidney failure and oral candidiasis were identified. The duration of elevated risk varied by condition, ranging from a week (eg, rhabdomyolysis) to 5 weeks (eg, hemoptysis) or 8 weeks (eg, chronic bronchitis).
CONCLUSIONS: We found that influenza is associated with conditions not typically considered. Also, some symptoms last longer than commonly appreciated, suggesting that some patients may suffer from a prolonged illness due to the influenza infection. Our findings have important implications for not only demonstrating the full burden of influenza but also informing future studies to prevent or treat influenza.