Influenza season 2017/2018

The 2017/2018 influenza season

The 2017/18 influenza season was unusual in terms of prevalence, duration and types of influenza virus in circulation. The season was prolonged and characterised by a very high level of influenza activity. The high level of activity was reflected in a high number of patients admitted to hospital with influenza and a substantial excess mortality among both adults and elderly people. The dominant virus was influenza B of the Yamagata line. It is rare for influenza B to be dominant, and this type is also not normally associated with serious disease that requires hospitalisation or with excess mortality among elderly people. In the final part of the season there was also a considerable number of influenza A cases.

Additionally, in this season, the influenza B virus included in the year’s trivalent vaccine and the influenza B virus in circulation were different, causing a low influenza B vaccine effectiveness.

Influenza monitoring methods

From Week 40, 2017 to Week 20, 2018, the occurrence of influenza-like illness (ILI) was reported to the sentinel surveillance by an average of 90 GPs weekly, compared with 101 in the previous season. The ILI occurrence was also followed daily via the nationwide surveillance performed by emergency call service physicians and also via the 1813 Helpline in the Capital Region of Denmark.

ILI in the population was monitored through Influmeter; a web-based monitoring system in which anyone residing in Denmark can report influenza-like symptoms independently of any healthcare contact. More information about Influmeter.

The occurrence of laboratory-verified influenza was monitored via the Danish Microbiology Database (MiBa), and the occurrence and any changes in the influenza virus were monitored via samples submitted to the National Danish WHO Influenza Centre at Statens Serum Institut (SSI).

Influenza-like admissions are monitored through weekly data capture from existing registers. Patients diagnosed with influenza are identified in the MiBa, and information on admission to hospital, intensive care therapy and underlying condition is collected from the National Patient Register. Furthermore, vaccination information is collected from the Danish Vaccination Register and information on fatalities from the Danish Civil Registration Register (the CPR Register). Corresponding information is used for the calculation of vaccine effectiveness. Mortality in the population in general is assessed through weekly collection of information on deaths from the CPR Register.

Occurrence of influenza-like illness

The share of patients who contacted their GP for ILI was increased from week 52, 2017, and continued to be so up until Week 14, 2018. In Week 5, 2018, the activity in general practice reached the intermediate level, and by Week 7 activity was at a very high level with a consultation percentage of 4.3. The activity remained high in Week 8, but then receded to intermediate and low levels in the following weeks. It is known that the consultation percentage in general practice increases artificially in the weeks of the winter holidays, as GPs have fewer planned consultations in these weeks. 

Measured by the sentinel surveillance scheme, the influenza activity was higher than in the four previous seasons, but it did not reach the high level recorded in 2012/13, which was dominated by influenza A(H3N2), Figure 1. The consultation percentage recorded by the emergency call service physicians was higher than the one recorded the previous year, but not higher than the one registered for the 2015/16 or the 2012/13 seasons, Figure 2.





Among Influmeter participants, the same seasonal variation in ILI was seen in 2017/18 as in the preceding years, Figure 3. A total of 331 ILI cases were reported among the season’s 1,218 Influmeter participants. The highest number was observed in Week 5 when 5% reported ILI.

Laboratory diagnostics MiBa

MiBa data extraction showed that a total of 56,113 patients were tested for influenza from Week 40, 2017, to Week 20, 2018, Figure 4. In all, 16,093 hereof tested positive to influenza (29% of those tested); 4,554 to influenza A (8%) and 11,539 to influenza B (21%), Figure 4. Season 2017/18 was thus the season with the highest number of influenza samples taken since 2010, and the share of positive samples was also higher than in previous seasons, Figure 5.

Laboratory surveillance at Statens Serum Institut

In the course of the season, the National Danish WHO Influenza Centre at the SSI tested samples from a total of 5,940 patients, including 471 patient samples submitted by the sentinel physicians, 173 diagnostic patient samples submitted for diagnostics by GPs and other laboratories, and surveillance samples from 5,296 patients submitted by departments of clinical microbiology. Among these, 5,516 (93%) patients tested positive to influenza virus of type A or B. The distribution of patients by identified virus strains is presented in Table 1 (10% of the 5,516 detected influenza viruses were not sub-typed). A total of 68% of the influenza-positive patients had influenza B, 17% had influenza A(H3N2) and 14% had influenza A (H1N1)pdm09.

Testing of 644 airway samples for 20 different airway viruses identified virus in 403 samples (63%): including 248 (62%) that were positive to influenza and 155 (38%) that tested positive to other airway viruses. Rhinovirus (10%) and respiratory syncytial virus (8%) were the second-most frequently observed group of viruses followed by human metapneumovirus (5%), coronavirus (4.5%), parainfluenza (4%), adenovirus (3%), enterovirus (3%) and parechovirus (0.5%), Figure 6.

Antiviral resistance

The national responsibility for testing influenza samples for antiviral resistance among patients who initiate antiviral treatment lies with the National Danish WHO Influenza Centre located at the SSI.

In all, 71 samples were received from 24 patients in the 2017/18 season, and analyses are still ongoing. Additionally, in the 2017/18 season, a total of 216 randomly selected samples were tested for resistance to the neuroaminidase inhibitors oseltamivir (Tamiflu®) and zanamivir (Relenza); no samples were resistant to both of these antiviral agents. The presently circulating influenza viruses are all naturally resistant to the ion channel inhibitors Amantadine and Rimantadine.

Admissions and treatment at intensive care departments

From week 40, 2017 through week 20, 2018, a total of 7,667 patients with laboratory-confirmed influenza were admitted to hospital according to the National Patient Register, Figure 7. This is the highest number recorded since 2010/11, which is the period for which we have data on laboratory-verified influenza admissions to hospital. A total of 2,399 had influenza A (31%), 5,306 had influenza B (69%) and 37 patients had a double infection. One patient had influenza C. The median admission age was 69 years for all admitted influenza patients, but considerable variation was present as patients admitted with influenza A(H1N1), A(H3N2) and influenza B had a median age of 52, 74 and 70 years, respectively. In all, 95% had underlying risk factors for a serious influenza course. The most frequently reported risk factors were chronic pulmonary disease/asthma (5,630 patients, 73%) age ≥65 years (4,451 patients, 58%) and cardiovascular disease (3,589 patients, 47%). For 412 patients (5%), information on risk factors was not available. A total of 603 (7.9%) patients died within 30 days after the latest positive influenza test. This share is in line with the one recorded in previous years. Even so, the majority of the admitted patients (4,835, 63%) were not registered as having received influenza vaccination.

A total of 549 (7%) of the hospitalised patients received intensive care, which is the same share as in the previous season and lower than in the 2015/16 season, which was the latest season to record many influenza B cases, Figure 8. Of the 549 patients, 199 (36%) had influenza A detected and 351 had influenza B detected (64%). Of these, 48% were women and 52% men. The median age was 68 years. A total of 99% had underlying risk factors for a serious influenza course. The most frequently occurring risk factors among the admitted patients was chronic pulmonary disease/asthma (489 patients, 89%), cardiovascular disease (340 patients, 62%) and age ≥65 years (335 patients, 61%). For 6 patients (1%), information on risk factors was not available. A total of 167 (30%) patients died within 30 days after the latest positive influenza test.

Mortality

Citizens may die from influenza without ever coming into contact with Danish healthcare to have their influenza detected. This may, e.g., occur among nursing home residents. Therefore, the SSI estimates excess mortality in the general population using statistical models that are based on historical data (excess mortality = observed mortality minus expected mortality). Additionally, by comparing excess mortality with information about influenza activity, it can be assessed how big a part of the excess mortality may be attributed to influenza.

In the course of the year’s influenza season, a significantly increased influenza-related mortality was recorded from Week 9 to Week 15, 2018. The total estimated excess mortality related to influenza for the entire 2017/18 season was 1,644 deaths, Table 2. This is the highest total influenza-related mortality observed in any season since 2010/11, which is the period comprised by our analysis. Influenza-related mortality varies from one age group to the next and between seasons depending on the circulating influenza viruses in the respective years. Among adults aged 15-64 years and elderly aged 65 years or above, 2017/18 recorded the highest mortality in the entire period comprised by the analysis, whereas younger age groups have seen higher mortalities.

Mortality in Europe is monitored through the joint European programme, EuroMOMO (www.euromomo.eu), provided by the SSI. On a weekly basis, the programme collects mortality data for the 24 participating European countries. In line with Denmark, most European countries recorded a considerable excess mortality in the 2017/18 season.

Vaccination coverage

Seasonal influenza vaccination was offered free of charge to specific groups from 1 October through 31 December 2017 and also to pregnant women in their second and third trimesters and to persons with immunodeficiency and their contacts until 1 March 2018, EPI-NEWS 38/17.

According to information from the Danish Vaccination Register (DVR), a total of 702,300 risk-group citizens were vaccinated, including 554,200 citizens aged 65 years or more (including citizens who received vaccination at their workplace), 111,600 chronically ill persons, 20,000 early retirement pensioners, 13,400 pregnant women and 3,100 household contacts to immunodeficient people.

Slightly more vaccinations were given to all risk groups than in the 2016/17 season. The vaccination overage among elderly people aged 65 years and above increased from 47% in 2016/17 to 50% in 2017/18. A contributing cause is the statutory registration of vaccinations in the DVR, which means that vaccinations given to persons in this age group, e.g. at work sites, now also appear from the register. Nearly 67,400 vaccinations were recorded as having been given to persons below 65 years of age who do not belong to a risk group. Data on the number of influenza vaccinations given to risk group persons and the vaccination coverage among elderly persons above the age of 65 years distributed by area and municipality are available at www.ssi.dk/data.

Influenza vaccine effectiveness

By linking information about samples tested for influenza from the MiBa to vaccination data from the Danish Vaccination Register and data on influenza type from the National Influenza Centre, the SSI could calculate the influenza vaccine effectiveness (VE) in a case-control study. The influenza A (H3N2) vaccine effectiveness among persons aged 65 years and above was 0%, and among 15-64-year-olds it was 13% (95% confidence interval: -20, 37). The low vaccine effectiveness compared with influenza A (H3N2) was expected because the circulating viruses had changed in relation to the virus strain in the vaccine. The influenza A (H1N1) vaccine effectiveness among persons aged 65 years and above was 45% (95% confidence interval: 28-58); among the 15-64-year-olds, it was 50% (95% confidence interval: 35-62). Compared with influenza B among persons aged 65 years and above, the vaccine effectiveness was 30% (95% confidence interval: 24-35), and among the 15-64-year-olds it was 33% (95% confidence interval: 26-39). The influenza B vaccine effectiveness was lower than in previous years because the influenza B type circulating in the population was not included in the vaccine. The observed vaccine effectiveness against influenza B may possibly be owed to cross-protection. The estimates are adjusted for age, sex, underlying conditions, calendar month during which the patient was tested and for the patient having been admitted to hospital or not.

Influenza vaccine 2018-2019

The WHO's recommendation for next year's seasonal vaccine is as follows:

- A/Michigan/45/2015 (H1N1) pdm09-like virus
- A/SingaporeINFIMH-16-0019/2016 (H3N2)-like virus (NEW VIRUS)
- B/Colorado/06/2017-like virus (Victoria line)

For the four-valent influenza vaccines, the recommendation is to include another influenza B virus of the type

- B/Phuket/3073/2013-like virus (the B/Yamagata line)

Commentary

The Danish 2017/18 influenza season was tough, and many citizens were affected by influenza, particularly influenza B. The consequence was more hospitalisations and more deaths than were recorded for previous influenza seasons. A similar image was reported for the remaining parts of Europe, as a prolonged season dominated by influenza B produced a high number of hospitalisations and an elevated mortality in many countries. Influenza B is normally described as a seasonal influenza virus that mainly affects school children and young adults and which is not known for any association with excess mortality among the elderly, in contrast to, e.g., influenza (H3N2). This may possibly be so because influenza B virus does not mutate quite as rapidly as influenza A, why immunity may increase with age. This season has therefore been unusual as many older persons were also affected by influenza B. Thus, the median age for people admitted to hospital with influenza was 70 years of age.

The vast majority (95%) of the patients who were hospitalised with influenza belonged to a risk group for serious influenza, i.e. had a chronic disease or were older than 65 years of age. This underlines the importance of preventing influenza in these groups through vaccination. Nevertheless, the majority of the admitted patients (63%) had not received influenza vaccination. Even though the trivalent vaccine has enjoyed a lower than expected effectiveness this year to both influenza A(H3N2) and influenza B, vaccination remains the best form of influenza prevention that we have.

Even at a low effectiveness level, vaccination prevents both hospitalisations and deaths. Furthermore, the disease often runs a less serious course if you do fall ill anyway. On the positive side, it is worth noting that the vaccination coverage increased to 50% among the elderly who are 65 years or more. This percentage, however, is still far from the WHO objective of a 75% vaccination coverage among elderly who are 65 years of age or more.

According to the WHO, next year’s trivalent influenza vaccine will also not contain influenza B/Yamagata. This is so because the virus is not expected to circulate in the same manner during the next season as a considerable share of the population will be immune following the massive spreading of infection in the current season.

This season saw a record-high number of microbiologic samples tested for influenza, and the positive percentage was generally high. This is so because many have fallen ill, but part of the explanation may also be owed to the fact that several hospital departments have introduced so-called rapid tests or “point-of-care tests”. That means that access to influenza test results at Danish hospital departments is now more rapid and less complicated. This is a good thing as a diagnosis and, if needed, isolation, can be achieved more rapidly. Increased testing may have contributed to the fact that this year has recorded more hospitalisations following the detection of influenza than previous seasons. Nevertheless, this year also recorded more influenza patients receiving intensive therapy than in previous seasons, and this is a patient group that we already expect will be tested thoroughly.

(The SSI Influenza Team: I.G Helmuth, T.G. Krause, L.K. Knudsen, H-D. Emborg, J. Nielsen, L.S. Vestergaard, C. Kjelsø, H. Bang, Infectious Disease Epidemiology and Prevention, R. Trebbien, T.K. Fischer, Virology Surveillance and Research)