Dominant influenza A(H3N2) and B/Yamagata virus circulation in EU/EEA, 2016/17 and 2017/18 seasons, respectively

Cornelia Adlhoch1, René Snacken1, Angeliki Melidou2, Silviu Ionescu1, Pasi Penttinen1, the European Influenza Surveillance Network3 1. European Centre for Disease Prevention and Control (ECDC), Stockholm, Sweden 2. Microbiology Department, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece 3. The members of the European Influenza Surveillance Network are listed at the end of article

We use surveillance data to describe influenza A and B virus circulation over two consecutive seasons with excess all-cause mortality in Europe, especially in people aged 60 years and older.Influenza A(H3N2) virus dominated in 2016/17 and B/Yamagata in 2017/18.The latter season was prolonged with positivity rates above 50% among sentinel detections for at least 12 weeks.With a current west-east geographical spread, high influenza activity might still be expected in eastern Europe.
The yearly influenza epidemics during each winter season vary in burden and severity.During the 2016/17 and 2017/18 seasons, all-cause excess mortality was observed during periods of high influenza virus circulation [1,2].Our aim is to describe and compare the pattern of influenza virus circulation and related disease severity by number of patients and fatal cases in intensive care units (ICUs) across European Union/ European Economic Area (EU/EEA) countries for the seasons 2016/17 and 2017/18.As influenza circulation progressed from a west to east direction across Europe in 2017/18, a better understanding of the current epidemiological situation might help to prepare countries in the eastern part of the World Health Organization (WHO) European Region for high influenza activity and severity [3].

Influenza surveillance in Europe
The European Influenza Surveillance Network (EISN) performs influenza surveillance from week 40 to week 20 of the following year [4,5].Weekly epidemiological and virological influenza data are collected from 30 EU/EAA countries and 11 countries report data on severe and fatal cases with laboratory-confirmed influenza in ICUs.Collected data in selected primary care settings include the percentage of these sentinel specimens testing positive for influenza [6].

Influenza virus circulation in the 2016/17 and 2017/18 seasons
The influenza virus positivity rates among sentinel specimens in EU/EEA countries over the 2016/17 and 2017/18 seasons are shown in Figure 1.

Results from surveillance in intensive care units
During 2016/17, 10 countries (Czech Republic, Denmark, Finland, France, Ireland, Romania, Slovakia, Spain, Sweden, and the United Kingdom (UK)) reported a total of 3,959 patients admitted to ICUs with laboratory-confirmed influenza (Figure 3).Of these patients, 96% (n = 3,813) were infected with influenza A virus, with 92% (1,465/1,592) of the subtyped A viruses being

Sentinel surveillance outcome
During the 2017/18 season, the influenza positivity of tested sentinel specimens passed the 10% threshold in week 47 2017, and the 50% threshold in week 52 2017.Positivity rates exceeded 52% for over 12 weeks indicating an unusually protracted peak period of influenza activity across the countries (maximum positivity: 61%; Figure 1).In the previous seven seasons, the number of weeks with a positivity of 50% and higher ranged between 0 (2009/10; 2013/  4).

Results from surveillance in intensive care units
Up to week 11 2018, 7,789 laboratory-confirmed influenza cases in ICUs were reported from 10 countries (Czech Republic, Denmark, Finland, France, Ireland, the Netherlands, Romania, Spain, Sweden, and the UK).Altogether, the UK (n = 2,983), France (n = 2,614) and Spain (n = 1,119) reported the vast majority of cases (86%; 6,716/7,789).Comparing data up to week 11 in 2017/18 to the complete 2016/17 season, this is an increase of ICU cases by 169%, 78% and 82% in these countries, respectively (Figure 3, Figure 5).
Influenza A was detected in 52% (n = 4,036) of the  [11,12].B viruses are described as causing milder disease and affecting more the younger age groups [13].In both seasons described here, however, we observed the highest number of severe cases in patients admitted to ICUs aged 60 years and older.As seen previously, patients infected with A(H1N1)pdm09 in ICU were slightly younger than patients infected with influenza A(H3N2) or B [14,15].
The reason for the prolonged and increased activity as well as the severe clinical impact of influenza B viruses during the 2017/18 season is not fully understood.
Countries perform either sentinel or universal hospital surveillance and no major change in reporting has occurred which would explain the increase of cases in ICU [16].
Although the most commonly used trivalent vaccine contained the B/Victoria lineage in both 2016/17 and 2017/18 seasons, a moderate vaccine effectiveness for B/Yamagata and A(H1N1)pdm09 in 2017/18 was observed; the effectiveness for A(H3N2) viruses in both seasons, however, was low [12,17].The moderate and low vaccine effectiveness as well as low vaccine coverage might have contributed to the lower protection in the population together with an accumulation of susceptible people since the last dominant B/Yamagata circulation five seasons ago [18].Further in-depth analyses are needed to describe the mostly affected population and identify relevant underlying co-morbidities and other factors, e.g.cold weather, that might have contributed to the prolonged virus activity and severity.Neuraminidase inhibitors remain an option for the prophylaxis and treatment of the currently circulating influenza viruses, and their use should be considered along the national and international guidance and recommendations, especially for cases with severe and rapidly progressing disease [19].The season in Europe has progressed in a marked west-east direction and countries in the eastern part of the WHO European Region should be prepared for possible cases of severe disease and impact on healthcare services.However, influenza activity remains high also in central and western parts of Europe with continuously observed all-cause mortality.

Figure
Figure 1Percentage positivity for influenza viruses among sentinel specimens, by week and season, European Union/European Economic Area, 2016/17 and 2017/18 (up to week 11)

1
Percentage positivity for influenza viruses among sentinel specimens, by week and season, European Union/European Economic Area, 2016/17 and 2017/18 (up to week 11) Distribution of laboratory-confirmed influenza cases admitted to ICU, by age group and type/subtype, in the reporting European Union/European Economic Area countries, 2016/17 season (n = 2,827 patients) a Figure 2 Distribution of viral types/subtypes in sentinel specimens by country, European Union/European Economic Area, influenza season 2016/17 EU/EEA: European Union/European Economic Area. a Only sentinel specimens are included.ICU: intensive care unit.aOf the total reported 3,959 patients, 2,827 where age information was available are included in the figure.Figure 4 Distribution of viral types/subtypes in sentinel specimens by country, European Union/European Economic Area, influenza season 2017/18 (up to week 11) EU/EEA: European Union/European Economic Area. a Only sentinel specimens are included.A(H3N2).The majority of cases (81%; 3,194/3,959) were reported from three countries, namely France (n = 1,469 cases), the UK (n = 1,109) and Spain (n = 616).Most cases in the 2016/17 season occurred up to week 11, by which 97% (3,837/3,959) of all the season's cases in ICU and 99% (593/599) of all fatal cases had been reported.
Distribution of laboratory-confirmed influenza cases admitted to ICU, by age group and type/subtype, in the reporting European Union/European Economic Area countries, 2017/18 season (up to week 11) (n = 4,754) a a Of the total reported 7,789 patients, 4,754 where age information was available are included in the figure.
The severity of disease caused by B viruses during the 2017/18 season is also reflected in the high proportion of B infections among fatal outcomes reported from ICUs (49%; 420/851).This is in contrast with the previous season in which 1% (5/599) of fatal cases in ICUs died of influenza B infection.In 2016/17, 0.5% (3/599) of the fatal cases were infected with A(H1N1)pdm09, 44% (266/599) with A(H3N2) and 54% (325/599) with