The Health Protection Surveillance Centre (HPSC) is working in collaboration with the National Virus Reference Laboratory (NVRL), the Irish College of General Practitioners (ICGP), the Departments of Public Health and the Intensive Care Society of Ireland (ICSI) on influenza surveillance in Ireland. Influenza activity during the 2010/2011 influenza season was regarded as moderate. Influenza A (H1) 2009 was the predominant influenza virus circulating for most of the season, with influenza B dominating in the latter part.
Methods
Sixty sentinel general practices (located in all HSE-Areas) reported electronically to ICGP the number of patients who consulted with influenza-like illness (ILI)1 on a weekly basis. Sentinel GPs were requested to send a combined nasal and throat swab to the NVRL on one to two ILI patients per week. HPSC collated both clinical and virological sentinel GP data and also non-sentinel (specimens mainly from hospitals) virological NVRL data. Other surveillance systems set up to monitor ILI/influenza activity include a network of sentinel hospitals reporting admissions data, sentinel schools reporting absenteeism and enhanced surveillance of hospitalised influenza cases aged 0-14 years. Hospital admissions and school absenteeism data act as a crude indicator for influenza activity.
Several surveillance projects that were initiated/augmented during the 2009 influenza pandemic were continued during the 2010/2011 influenza season:
- Surveillance of all calls to GP out-of-hours (OOHs) centres were monitored for self-reported influenza. These data were provided by HSE-NE.
- Surveillance of all confirmed influenza notifications, including hospitalisation status reported to the Computerised Infectious Disease Reporting System (CIDR) in Ireland.
- Surveillance of all confirmed influenza adult and paediatric cases admitted to critical care.
- Enhanced surveillance of all confirmed influenza deaths.
Results
Sentinel GP Clinical Data
Influenza activity in Ireland was high during the 2010/2011-influenza season, peaking during week 1 2011 (early January) at 202.1 per 100,000 population. This is the highest recorded ILI rate since influenza surveillance began in 2000 (figure 1). ILI rates first increased above baseline levels (17.8 per 100,000) during week 50 2010 (December) and remained there for 12 weeks until week 10 2011 (March). The highest age specific ILI rates were in the 5-14 year age group (peaking at 254.1/100,000), followed by those aged 15-64 years (244.9/100,000), 0-4 year olds (193.6/100,000) and those aged 65 years or older (66.0/100,000). ILI rates in those aged less than 15 years were lower than during the pandemic period, however ILI rates in the 15-64 year age group and those aged 65 years or older were higher.
Figure 1: ILI sentinel GP consultation rates per 100,000 population, baseline ILI threshold rate, and number of positive influenza A and B specimens tested by the NVRL2, by influenza week and season.
Source: Clinical ILI data from ICGP and virological data from the NVRL.
The NVRL tested 1054 sentinel specimens for influenza virus during the 2010/2011 season. Almost half of all sentinel specimens (513; 48.7%) were positive for influenza: 279 influenza A (267 A (H1) 2009, 9 A (H3) and 3 A unsubtyped) and 234 influenza B. At the peak of influenza activity, the proportion of influenza positive sentinel specimens reached 72.4% (during week 1 2011).
The NVRL tested 7,114 non-sentinel respiratory specimens during the 2010/2011 season, over one fifth (1518; 21.3%) of which were positive for influenza: 1157 influenza A (1099 A (H1) 2009, 31 A (H3) and 27 A unsubtyped) and 361 influenza B. Eight influenza A cases were co-infected with influenza B: 7 with influenza A (H1) 2009 and one with influenza A (unsubtyped). One influenza B case was also co-infected with respiratory syncytial virus (RSV).
Influenza A (H1) 2009 was the dominant influenza type/subtype detected during the first half of the season, with influenza B dominating during the latter half. Influenza A accounted for 70.7% of all influenza positive specimens and influenza B for 29.3% during the 2010/2011 season. Influenza A (H1) 2009, accounted for 67.3% of all positive influenza specimens.
Antiviral resistance & antigenic characterisation
The NVRL tested eight non-sentinel specimens from six confirmed influenza A (H1) 2009 cases for antiviral resistance. All six patients were hospitalised and admitted to critical care. One (12.5%) of the eight specimens tested was resistant to oseltamivir, carrying the H275Y mutation. As part of the WHO Global Influenza Surveillance Programme, a proportion of influenza viruses (10 A (H1N1) 2009 and 2 B viruses) circulating in Ireland during the 2010/2011 season were submitted to the WHO Collaborating Centre for Reference and Research on Influenza (Mill Hill, London) for characterisation. Antigenic characterisation results for the circulating influenza A (H1N1 2009) and for the influenza B isolates showed good reactivity with the 2010/2011 influenza vaccine strains, A/California/7/2009 and B/Brisbane/60/2008, respectively. Therefore, indicating a good match between the circulating and vaccine strains.
Outbreaks, GP OOHs, Sentinel hospital & school data
Fourteen general ILI/influenza outbreaks were reported to HPSC: eight ILI outbreaks, five influenza A (H1) 2009 outbreaks and one outbreak associated with both influenza A (H1) 2009 and influenza B. Five outbreaks were reported from HSE-E, seven from HSE-S and two from HSE-W. Two outbreaks were in healthcare settings (one of which was a maternity hospital), seven in schools, one in a community setting, one in a residential institution, one in a prison, one travel related outbreak and one outbreak reported as ‘Other’ setting.
The percentage of influenza-related calls to GP out-of-hours services in Ireland, peaked during week 1 2011 at 14.7%. This is higher than the proportion recorded during the pandemic period. During the peak of activity, each service received on average five calls per hour relating to influenza.
Hospital respiratory admissions in sentinel hospitals peaked during week 52 2010 (figure 2), one week prior to the peak in sentinel GP ILI consultation rates. Sentinel school absenteeism data are not presented in this report, as most sentinel schools were closed for an unusually extended period of time in December and January due to severe weather and road conditions, coinciding with the peak of influenza activity.
Figure 2: Total hospital respiratory admissions in nine sentinel hospitals and ILI sentinel GP consultation rate per 100,000 population by week for the 2010/2011 influenza season.
It should be noted that admissions data from one sentinel hospital were not available for weeks 40-50 2010.
Influenza notifications
A total of 2233 influenza notifications were reported on CIDR during the 2010/2011 influenza season. The peak of influenza notifications occurred during week 2 2011, one week following the peak in ILI consultation rates and GP OOHs flu calls. Of the 2233 notifications, 1324 (59.3%) were influenza A (H1) 2009, 23 (1.0%) were influenza A (H3), 203 (9.1%) were influenza A (unsubtyped) and 683 (30.6%) were influenza B.
Hospitalisation
Nine-hundred and forty-five cases with confirmed influenza were hospitalised during the 2010/2011 influenza season. Similar to the pandemic period, the highest age specific rate in hospitalised cases was in the 0-4 year age group (61.9 per 100,000 population) (table 1). Of the 945 hospitalised cases, 602 (63.7%) were influenza A (H1) 2009, 7 (0.7%) were influenza A (H3), 109 (11.5%) were influenza A (unsubtyped) and 227 (24.0%) were influenza B.
Table 1: Age specific rate for confirmed influenza cases hospitalised and admitted to critical care during the 2010/2011 influenza season.
Age specific rates are based on the 2006 CSO population census.
Pregnancy
A total of 81 laboratory confirmed influenza cases were reported as pregnant during the 2010/2011 season. Fifty-one (63.0%) of these cases were hospitalised: influenza A (H1) 2009 was detected from 42 (82.3%) of these cases, influenza A (unsubtyped) from two cases and influenza B from seven cases. Eight (15.7%) of all reported hospitalised pregnant cases were admitted to ICU, one of whom died.
Enhanced surveillance hospital data on 0-14 year age group
A total of 547 confirmed influenza cases aged between 0 and 14 years were notified on CIDR for the season, 50.3% (275) of whom were hospitalised. Enhanced surveillance data were available on almost all of the hospitalised cases (266, 96.7%). Of these cases, almost two-thirds (174, 65.4%) were positive for influenza A (144 influenza A (H1) 2009, 1 A (H3) and 29 A (unsubtyped)) and over one-third (92, 34.6%) were positive for influenza B. The predominant influenza type/subtype was influenza A (H1) 2009, accounting for 54.1% of all cases in this age group. The median age of cases was 2 years, ranging from 6 days to 14 years. The majority, 66.5% were aged between 0 and 4 years. The most frequently reported symptoms included: fever (74.8%), cough (65.8%) and gastrointestinal symptoms (33.8%). The most frequently reported complications included: secondary bacterial pneumonia, primary influenza viral pneumonia, bronchitis and other respiratory complications, such as bronchiolitis and chest infections. The median length of stay in hospital was 3 days (ranging from 1 - 35 days). Over one-third (81, 35.4%, n=229) of cases had underlying medical conditions. Chronic respiratory disease was the most frequently reported underlying medical condition (33/229, 14.4%). Fourteen (6.1%) cases had more than one underlying medical condition. Vaccination status was known for over two-thirds (175, 65.8%) of children. Only five of 175 (2.9%) cases were vaccinated with the 2010/2011 influenza vaccine. Three of the five vaccinated cases had also received the pandemic influenza vaccine. Anti-viral treatment was commenced on 38% (73/194) of cases. Fourteen percent of cases were associated with an ILI/influenza outbreak. Fourteen cases were admitted to critical care and two cases died.
Critical care
Of the 945 hospitalised cases, 121 (12.8%) were admitted to critical care (107 adults and 14 paediatric cases). The highest age specific rate for patients admitted to ICU was in the 55-64 year age group (5.9 per 100,000 population) (table 1). The median age of paediatric cases was one year of age and the median age of adult cases was 51 years. Eight-one (81/107, 75.7%) adults and nine (9/14, 64.3%) paediatric cases had pre-existing medical conditions.3 The most frequently reported underlying medical conditions for adults included chronic respiratory disease (n=42/107, 39.3%) and chronic heart disease (n=26/107, 24.3%). The most frequently reported underlying medical conditions for paediatric cases included chronic neurological disease (n=4/14, 28.6%) and chronic respiratory disease (n=3/14, 21.4%). Nine (64.3%) paediatric and 98 (91.6%) adults were ventilated during their stay in ICU. The median length of stay in ICU for paediatric cases was 6.5 days (ranging from 1 - 20 days) and for adult cases was 13.5 days (ranging from 1 - 135 days). Eighteen (n=121, 14.9%) cases were vaccinated during the 2010/2011 influenza season. Clinical outcome data are presented below.
Mortality data
During the 2010/2011 influenza season, 38 influenza-associated deaths4 were reported, compared to 295 during the pandemic period. The median age of cases who died during the 2010/2011 influenza season was 57 years, ranging from 2 – 83 years. Thirty-two (84.2%) cases had underlying medical conditions. Thirty-two (84.2%) cases were admitted to ICU. Of the 38 cases who died, 32 were infected with influenza A (H1) 2009, one with influenza A (H3), one with influenza A (unsubtyped), one co-infection of influenza A (H1) 2009 and influenza B and three with influenza B. Vaccination status was known for 33 of the 38 (86.8%) deaths. The majority of cases (88%, 29/33) were not vaccinated with the 2010/2011 influenza vaccine. A summary table of confirmed influenza hospitalised and critical care cases and influenza-associated deaths for all ages is detailed in table 2.
Table 2: Summary table of confirmed influenza hospitalised and critical care cases and influenza-associated deaths for all ages.
It should be noted that risk factor data was not available for all age groups for the 2010/2011 season. Rates are based on the 2006 CSO population census.
Discussion
Based on all influenza/ILI data available at HPSC, influenza activity during the 2010/2011 was regarded as moderate. The highest ILI rates were reported since the GP sentinel surveillance scheme began (in 2000). ILI/influenza activity was very intense during the peak of influenza activity (week 1 2011). The predominant circulating influenza virus was influenza A (H1) 2009, however during the latter part of the season, influenza B dominated. Fewer cases were admitted to hospital than during the pandemic period, however more cases were admitted to critical care and more influenza-associated deaths were reported. The median age of severe cases was older compared to the pandemic period. Similarly, the proportion of severe cases with underlying medical conditions, although still high, was less than that reported during the pandemic period. Critical care and hospitalisation rates were similar to the pandemic period. Activity was more intense in hospitals and critical care units during the peak of the 2010/2011 influenza season than during the peak of the pandemic period. This would indicate an increased burden on hospitals and critical care units during the Christmas and New Year period when the number of influenza admissions increased significantly. During the pandemic, activity in hospitals and critical care units was spread over a longer period of time but the intensity was not as marked at the peak of influenza activity.
It is notable that influenza vaccination rates in hospitalised children, patients in critical care units and among confirmed influenza deaths was low at 2.9%, 14.9% and 12.1% respectively. A substantial number of these cases belonged to risk groups for influenza vaccine. This indicates that a stronger message regarding the benefits of the influenza vaccine in reducing significant illness and death need to be reiterated among health care professionals and the public.
For the 2011/2012 influenza season, existing surveillance systems have been strengthened and a number of additional measures have been put in place in Ireland to improve surveillance of ILI/influenza. At HPSC, initiatives have been implemented to streamline reporting and to capture additional data on influenza vaccination status, underlying medical conditions and antiviral treatment. The NVRL has improved standardisation of procedures, testing and reporting algorithms for characterising influenza viruses. They have also introduced multiplex PCR (swabs) for influenza A, B, RSV, adenovirus, parainfluenza virus types -1 and -3 and human metapneumovirus. The NVRL will also continue monitoring for oseltamivir resistance.
All influenza surveillance projects described for the 2010/2011 influenza season will continue for the 2011/2012 influenza season. A review of the influenza surveillance system in critical care units was conducted at the end of the 2010/2011 influenza season and improvements were made. A pilot project is being implemented in two critical care units to monitor morbidity and mortality from severe acute respiratory infections (SARI) for the 2011/2012 influenza season. Additional projects not detailed in this report include participation in a European influenza vaccine effectiveness study (I-MOVE project), seasonal influenza vaccine uptake in those aged 65 years or older and an all-cause mortality monitoring project associated with the European mortality monitoring group (EuroMOMO). Data from all of these surveillance systems will assist in guiding the management and control of influenza and any future epidemics or pandemics. www.hpsc.ie
References available upon request.
L Domegan, J O’Donnell, D O’Flanagan, HPSC; M Joyce, O Levis, ICGP; M Duffy, J Moran, S Coughlan, NVRL.
Acknowledgements
The authors would like to thank the sentinel GPs, ICGP, NVRL, Departments of Public Health, sentinel schools and hospitals, ICSI and HSE-NE for their contributions towards influenza surveillance throughout the influenza season. We would also like to thank all those at HPSC that have worked on influenza surveillance during the pandemic period and the 2010/2011 influenza season.
Notes
1. ILI is defined using the EU case definition which is sudden onset of symptoms AND at least one of the following four systemic symptoms: fever, malaise, headache, myalgia; AND at least one of the following three respiratory symptoms: cough, sore throat, shortness of breath.
2. In addition to the NVRL, Cork University Hospital (CUH) and Galway University Hospital(s) (GUH) also tested for influenza A (H1) 2009 during the pandemic period.
3. Some cases had more than one underlying condition.
4. Influenza-associated deaths include: deaths with influenza as the primary/main cause of death on the death certificate or as reported by the clinician as the cause of death or deaths with influenza listed as the cause of death anywhere on the death certificate.
5. This is an increase on previously reported deaths for the pandemic period, following searches of the General Registrar’s Office registered deaths data.
