Navigation and service

Report on the Epidemiology of Influenza in Germany 2017/2018

Executive Summary

full report in German

The information on the epidemiology of influenza in Germany for the 2017/18 season is mainly based on the analysis and assessment of data collected by the Robert Koch Institute’s (RKI) various surveillance systems for the monitoring of acute respiratory infections (ARI), particularly influenza, in Germany. The sentinel system of the Working Group on Influenza (AGI) with its syndromic surveillance of acute respiratory diseases and the virological surveillance of respiratory pathogens continues to be a central instrument in the overall concept of influenza surveillance in Germany. The virological data of the AGI on influenza are supplemented by results of six state laboratories cooperating with AGI in Baden-Württemberg, Bavaria, Mecklenburg-Western Pomerania, Saxony, Saxony-Anhalt and Thuringia. Mecklenburg-Western Pomerania contributed syndromic data from sentinel practices of the state’s own surveillance. The National Reference Center for Influenza (NIC) conducted additional virological analyses of circulating influenza viruses. Mandatory reports of laboratory confirmed cases of influenza were obtained from the German local health authorities who submitted notifications via state health authorities to the RKI. These were also included in the report, as were the results from the web-based participatory surveillance system GrippeWeb where 14,000 participants are registered. Finally we present results from the electronic module of the AGI (SEEDARE), the hospital surveillance system for severe acute respiratory infections (ICOSARI) and timely mortality data from Berlin.

This season we identified the first influenza (B) viruses within the AGI sentinel already in the 40th calendar week (CW) 2017. In the 52nd CW 2017 the proportion of influenza-positive samples (positivity rate) rose to 32 %. Thus, in the 2017/18 season the flu epidemic began in the 52nd CW 2017 and it ended in the 14th CW 2018 in early April. The activity of acute respiratory diseases, measured by the so called practice index, reached values of high ARI activity from CW 6 to 11 2018. Notably, the highest values of the practice index during the peak of the flu epidemic in the 8th to 10th CW 2018 were not reached in any of the seasons since 2001 when the RKI took over scientific leadership.

We estimate that a total of approximately 9 million influenza-attributable medically attended acute respiratory illnesses (IMAARI) occurred (95 % confidence interval (CI), 8 – 10 million). In all age groups the majority of IMAARI were caused by influenza B with the exception of the 0- to 4-year-old children where most IMAARI were caused by influenza A(H1N1)pdm09. Influenza led to an estimated 5.3 million (95 % CI, 5.1 – 5.5 million) physican-signed sick leave, or – among persons who are not in employment, such as children and elderly – to the need to stay home. The estimated number of influenza-related hospital admissions based on data from primary care practices was 45,000 (95 % CI, 42,000 – 47,000).

Compared to the severe seasons in 2012/13 and 2014/15 the number of IMAARI surpassed those seasons by 2 million. Estimates of influenzarelated hospital admissions were also higher than in 2012/13 and 2014/15. This season, for the first time, the number of laboratory-confirmed hospitalized cases of influenza that were reported to the local health authorities surpassed the number estimated by the AGI which, however, is based exclusively on hospital admissions from GP or pediatric practices, and does not include direct admissions to the hospital bypassing primary care practices. The estimates for IMAARI and influenza-related hospital admission showed, that in the 2017/18 season, especially those in the age groups 35 years and older were affected.

According to the virological sentinel surveillance conducted by the NIC influenza B viruses dominated the flu epidemic from the start. They were the most frequently detected with 68 %. Influenza A(H1N1)pdm09 viruses were identified in 28 % of influenza-positive samples, and influenza A(H3N2) viruses only in 4 %. 99 % of all influenza B viruses detected in the sentinel belonged to the Yamagata-lineage (B/Yam), while the trivalent vaccine for the 2017/18 season contained an influenza B component of the Victoria-lineage (B/Vic).

The Influenza B/Yam viruses and the A(H1N1)pdm09 viruses reacted well with post-infection ferret antiserum raised against the respective vaccine viruses, while the A(H3N2) and B/Vic viruses showed better agreement with the vaccine strains recommended for the upcoming 2018/19 season. The NIC also conducted molecular analysis of influenza-positive samples in the context of the investigation and management of influenza outbreaks in a kindergarten and two hospitals by local health authorities where also severe cases had occurred. In addition, samples were analysed that were sent to the NIC coming from other patients with a severe or fatal course. Finally, the NIC had also tested approximately 27 % of the influenza viruses for resistance against antivirals. Except for one influenza B virus, where a resistance-associated mutation of the neuraminidase was identified, all viruses tested were susceptible to the neuraminidase inhibitors oseltamivir and zanamivir.

The analysis of the GrippeWeb data shows that in the 2017/18 season in particular people with influenza-like-illness (ILI) aged 35- to 59-years old visited a GP substantially more frequently than in the three seasons before. Although the ILI rate measured in the population by GrippeWeb did not quite reach the levels of the 2014/15 and 2016/17 seasons at the peak, overall the period of increased ILI rates lasted longer than in 2016/17.

The distribution of the ICD-10 diagnostic codes for ARI in ambulatory care is shown in the more detailed analysis of the SEEDARE data. The number of consultations in which certain ICD-10 codes for upper respiratory tract infections, influenza or lower respiratory tract infections have been used, showed a clear seasonal pattern. The proportion of patients admitted to hospital was considerably higher in the subgroup of patients with ICD codes for influenza, pneumonia, and other lower respiratory tract infection than among patient with any ICD-10 code used for ARI. Using the case-based anonymous information from the SEEDARE module, further respiratory syndromes can also be specifically analysed, such as illnesses that have been coded as communityacquired pneumonia.

The German hospital surveillance system for severe acute respiratory infections (ICOSARI) contributed timely information that helped to evaluate the occurrence of severe influenza cases. In the 2017/18 season, the number of hospitalized SARI patients exceeded the number of cases in the three previous seasons. As was the case in the 2016/17 season, particularly elderly people with severe illnesses were hospitalized. Among SARI patients who needed intensive medical care, the age groups 35 years and older were particularly affected during the flu epidemic.

The exceptional severity of the 2017/18 season is also reflected in the analysis of excess mortality during the influenza period. The data shown in this report for Berlin with an estimated 1,100 additional deaths exceed the already high estimates for the 2016/17 season.

The World Health Organization’s (WHO) annual recommendations on influenza vaccines, the recommendations of the German Standing Committee on Vaccination (STIKO), and the assessment of the influenza vaccine effectiveness for the 2017/18 season are all presented in the chapter »Influenza Vaccination«. For the 2018/19 season, the WHO recommended a different composition of the trivalent influenza vaccine for the influenza A(H3N2) and influenza B (Victoria lineage) components in comparison to the Northern Hemisphere 2017/18 season:

  • an A/Michigan/45/2015(H1N1)pdm09-like virus (unchanged);
  • an A/Singapore/ INFIMH-16-0019 / 2016 (H3N2) -like virus (new)
  • a B/Colorado/06/2017-like virus (Victoria lineage) (new)

For quadrivalent vaccines containing the second B lineage (Yam), a B/Phuket/3073/2013-like virus (Yamagata lineage) was recommended, the same as in the 2017/18 season.

As in the previous seasons, the effectiveness of influenza vaccination in the 2017/18 season was assessed by analysing the virological surveillance data of the AGI. The vaccine effectiveness against laboratory-confirmed influenza B adjusted for sex, age group, presence of underlying disease and disease week was 1 % (95 % CI, -36 to 28 %). The effectiveness of the vaccine against laboratory confirmed influenza A(H1N1)pdm09 disease was 48 % (95 % CI, 11 to 70 %).

Lastly, the chapter on zoonotic influenza describes the situation on avian and porcine influenza in their respective animal species, as well as in humans.

As in previous years, no human case with zoonotic influenza virus infection was reported in Germany. However, also in the 2017/18 season, human infections with avian and porcine influenza viruses occurred worldwide. They were mostly attributed to exposure to infected animals. There is also no evidence of sustained human-to-human transmission with these zoonotic influenza viruses. As long as the influenza viruses circulate in livestock, sporadic human infections may continue to occur.

Date: 12.09.2018