Association between type-specific influenza circulation and incidence of severe laboratory-confirmed cases; which subtype is the most virulent?. (July 2020)

• Record Type: Journal Article
• Authors: Lytras, T.; Andreopoulou, A.; Gkolfinopoulou, K.; Mouratidou, E.; Tsiodras, S.
• Is Part Of: Clinical microbiology and infection. Volume 26:Number 7(2020)
• Page Start: 922
• Page End: 927
• Publication Date: 2020-07
• Subjects: Epidemiology -- Influenza -- Influenza complications -- Intensive care units -- Mortality; Medical microbiology -- Periodicals; Diagnostic microbiology -- Periodicals; Communicable diseases -- Periodicals; Infection -- Periodicals; 616.01
• DOI: 10.1016/j.cmi.2019.11.018
• Languages: English
• ISSNs: 1198-743X

Abstract

Abstract: Objectives: Excess population mortality during winter is most often associated with influenza A(H3N2), though susceptibility differs by age. We examined differences between influenza types/subtypes in their association with severe laboratory-confirmed cases, overall and by age group, to determine which type is the most virulent. Methods: We used nine seasons of comprehensive nationwide surveillance data from Greece (2010–2011 to 2018–2019) to examine the association, separately for influenza A(H1N1)pdm09, A(H3N2) and B, between the number of laboratory-confirmed severe cases (intensive care hospitalizations or deaths) per type/subtype and the overall type-specific circulation during the season (expressed as a cumulative incidence proxy). Quasi-Poisson models with identity link were used, and multiple imputation to handle missing influenza A subtype. Results: For the same level of viral circulation and across all ages, influenza A(H1N1)pdm09 was associated with twice as many intensive care hospitalizations as A(H3N2) (rate ratio (RR) 1.89, 95% CI 1.38–2.74) and three times more than influenza B (RR 3.27, 95%CI 2.54–4.20). Similar associations were observed for laboratory-confirmed deaths. A(H1N1)pdm09 affected adults over 40 years at similar rates, whereas A(H3N2) affected elderly people at a much higher rate than younger persons (≥65 vs. 40–64 years, RR for intensive care 5.42, 95% CI 3.45–8.65, and RR for death 6.19, 95%CI 4.05–9.38). Within the 40–64 years age group, A(H1N1)pdm09 was associated with an approximately five times higher rate of severe disease than both A(H3N2) and B. Discussion: Influenza A(H1N1)pdm09 is associated with many more severe laboratory-confirmed cases, likely due to a more typical clinical presentation and younger patient age, leading to more testing. A(H3N2) affects older people more, with cases less often recognized and confirmed.