Influenza B in the general population. Retrospective analysis of French surveillance data from 2003 to 2017

Context Much attention has been paid to influenza over the past three decades with a rare focus on influenza B. Since 2000, two distinct influenza B lineages (Yamagata and Victoria) circulate alternately or simultaneously, raising the issue of potential vaccine mismatch with trivalent vaccines that contain only one of the 2 B lineages. Since 2013, WHO influenza vaccine recommendations give advice on a vaccine strain for each influenza B lineage and manufacturers have developed quadrivalent seasonal vaccines containing the two lineages of influenza B viruses. In France, between 1984 and 2017, influenza surveillance in the general population has been performed by two sentinel networks, based on general practitioners and paediatricians, the Reseau national des GROG (from 1984 to 2014) and the Sentinelles network (since 2014). Practitioners of the network are providing weekly data describing their activity and collect nasal swabs in a fraction of their patients consulting for acute respiratory infection (ARI) or influenza-like infection (ILI). These swabs are sent by surface mail to the laboratories of the National Reference Centre. Objective Our study aims to describe the circulation of the two lineages of influenza B viruses in the general population, during fourteen influenza seasons in France (2003–2017). Methods A retrospective descriptive analysis of the GROG 2003–2013 database, completed by the Sentinelles surveillance data publicly available from 2014 was performed. Every virologically confirmed influenza positive case was included in the study, except for those with an influenza A and B co-infection. Patients whose age was unknown were also excluded. Results and discussion Overall, 21,070 virologically confirmed influenza cases could be included, comprising 5478 influenza B cases (26.0%). The contribution of influenza B to the seasonal influenza burden varied from year-to-year. Influenza B represented more than 5% of influenza viruses detected in eight of the fourteen studied seasons. Influenza B was considered as dominant (> 60% of all influenza viruses of the season) in 2005–2006 (61.8%) and 2015–2016 (71.5%) and was co-dominant (41–60% of all influenza viruses of the season) in 2010–2011 (47.4%) and 2012–2013 (55.0%). The influenza B viruses impact was mainly observed in children of the 5–14 years old group (33.0%) but was also significant in the ≥ 65 years old group (22.5%). In the 8 seasons where influenza B viruses substantially circulated, both Victoria and Yamagata lineages were detected. They respectively accounted for 57.8% and 42.2% of all influenza B cases for which the lineage was available. Each lineage was predominant during four seasons: Victoria in 2005–2006, 2008–2009, 2010–2011 and 2015–2016; Yamagata in 2004–2005, 2007–2008, 2012–2013 and 2014–2015. A mismatch between the dominant circulating influenza B lineage and the lineage included in the seasonal influenza vaccine was observed during four (28.5%) of the fourteen seasons analysed, including two seasons where influenza B viruses predominated (2005–2006, 2007–2008, 2008–2009 and 2015–2016). This is consistent with the results of a similar study carried out in 26 countries that reported a type B lineage mismatch in 25% of seasons. In our study, for 58.2% of the documented type B infections, the reported lineage was not included in the seasonal trivalent vaccine. Again, this is consistent with results recently published from Finland between 1999 and 2012 where 41.7% of the documented influenza B cases belonged to the mismatched lineage. Conclusion Our results support the added value of the quadrivalent vaccine to increase vaccine effectiveness.