Rapid communications Increase in invasive Streptococcus pyogenes and Streptococcus pneumoniae infections in England , December 2010 to January 2011

K Zakikhany1,2, M A Degail1,3, T Lamagni1, P Waight1, R Guy1, H Zhao1, A Efstratiou1, R Pebody1, R George1, M Ramsay (mary.ramsay@hpa.org.uk)1 1. Health Protection Agency (HPA), London, United Kingdom 2. The European Programme for Public Health Microbiology Training (EUPHEM), European Centre for Disease Prevention and Control (ECDC), Stockholm, Sweden 3. The European Programme for Intervention Epidemiology Training (EPIET), European Centre for Disease Prevention and Control (ECDC), Stockholm, Sweden

Increases in invasive Streptococcus pyogenes and S. pneumoniae above the seasonally expected levels are currently being seen in England.Preliminary analyses suggest that the high level of influenza activity seen this winter may be contributing to an increased risk of concurrent invasive bacterial and influenza infections in children and young adults.
Following the early and rapidly escalating start of the 2010/11 influenza season in England [1,2] the Health Protection Agency (HPA) became aware of a number of anecdotal reports of invasive bacterial infections complicating seasonal influenza or influenza A(H1N1) 2009.In parallel, analyses of routine surveillance data identified increases in Streptococcus pyogenes and S. pneumoniae infections [3,4].This triggered a cascaded alert from the United Kingdom (UK) Chief Medical Officer to healthcare professionals to be vigilant for bacterial co-infections complicating influenza cases [5].
The UK has experienced intense and widespread influenza activity this winter season due primarily to influenza A(H1N1)2009 virus with a significant contribution

Identification of cases
Cases of invasive S. pyogenes, S. pneumoniae, and Haemophilus influenzae infection (defined through the isolation of these organisms from a normally sterile site) were identified through isolate referral to the national or regional reference laboratories.Cases of meningococcal infection included those with a clinically compatible illness where an isolate of Neisseria meningitidis was referred or where meningococcal DNA was detected in a clinical specimen at the national reference laboratory.Confirmed infections due to Staphylococcus aureus and S. pyogenes were derived from reports to the HPA from laboratories in England.
Cases of influenza were defined as persons with influenza-like illness (ILI) with laboratory-confirmed influenza A or B infection reported by local or regional laboratories in England [6].
To obtain a minimum estimate of the potential importance of influenza as a risk factor for invasive bacterial infection, invasive bacterial surveillance data between 1 November 2010 and 14 January 2011 were matched on unique patient identifier (National Health Service (NHS) number, or name and date of birth if NHS number was unavailable) to laboratory-confirmed influenza diagnoses.Cases in both datasets with sample dates within two weeks of each other were considered as possible co-infections.

Invasive bacterial infections in England
When compared to the same period in the previous three years, surveillance data for 2010 and 2011 (1 July 2010 to 14 January 2011) do not indicate an overall increase in the number of invasive S. pneumoniae, S. aureus, H. influenzae or N. meningitidis infections (Table 1).In contrast, numbers of invasive S. pyogenes infections showed a slight elevation overall (Table 1), although more pronounced in December 2010 when 173 reports were received compared to an average of 99 for the same month in 2002-09 (range 68 to 147).
Increases in invasive S. pyogenes disease cases were noted in all age groups (Figure 1A) and were seen across all regions of England with the exception of Yorkshire and the Humber.Increases in invasive S. pneumoniae infections were seen exclusively in young adults (15-44 years) (Table 1), with numbers in December 2010 being much higher than the average for the same month of the previous three years (2007-09) (Figure 1B).
These increases coincided with increased influenza activity in December 2010, in particular in children (under 15 years-old) and young adults (15-44 years-old) (Figure 2).

Concurrent bacterial infections in seasonal influenza and influenza A (H1N1) 2009 cases in England
Linkage of influenza surveillance data to the 4,232 invasive bacterial surveillance records reported since the beginning of the 2011/11 influenza season (1 November 2010) to 14 January 2011, identified 144 (3.4%) cases co-infected with influenza (Table 2).Of the bacterial co-infections, the majority (85%) were diagnosed within the seven days after the date of laboratoryconfirmed influenza diagnosis (122/143).Around three quarters (109/143) of identified co-infections were influenza A, 26% (37/143) were influenza B and 2% had both infections.S. pyogenes and S. pneumoniae had the highest proportion of confirmed influenza coinfections compared to the other bacterial infections.Cases of S. pyogenes under the age of 15 years had the highest likelihood of influenza co-infection (14%) followed by cases aged between 15 and 44 years (13%).
Similarly, the highest proportion of co-infections with S. pneumoniae and influenza was found in the 15-44-year-olds (Table 2).

Discussion
Routine monitoring of surveillance data in England has identified a widespread increase in invasive S. pyogenes in December 2010 beyond the seasonally expected.A similar trend was not observed for other invasive bacterial pathogens where overall case numbers remained in line with previous seasons.Analysis of case fatality rates for all invasive bacterial pathogens studied were within the usual range.
Periodic upsurges in invasive S. pyogenes disease are reported by both European and non-European countries [7].The drivers behind these increases are not fully understood but are likely to reflect both natural cycles governed by population susceptibility and heightened transmission in specific risk groups (e.g.injecting drug users).Our preliminary findings suggest that the heightened influenza activity this season has contributed to an increased risk of invasive S. pyogenes infection in children and young adults as co-infections with S. pyogenes and influenza were specifically observed in these age groups.This is in line with incidence rates of influenza and influenza-like illness (ILI) which were highest in December 2010 in children ( and the subsequent change to a 13-valent conjugate vaccine in 2010 [9].Trends in older age groups, however, may have also been affected by recent changes in the vaccine programme.
As we were only able to match to laboratory-confirmed influenza, it is likely that we have underestimated the number of true cases of co-infections in the population, and influenza may be a more significant contributor to the overall rate of invasive infections.Furthermore, the importance of influenza as a risk factor for inva-

Table 1
Cases of invasive Streptococcus pyogenes, Streptococcus pneumoniae, Staphylococcus aureus, Haemophilus influenzae andNeisseria meningitidis infection diagnosed in England between 1 July and 14 January, 2007 to 2011

Table 2
[10]distribution of invasive bacterial infections with concurrent influenza A or B infection in England, 1 November 2010-14 January 2011 sive bacterial infection is likely to vary across different parts of the country[10].The changes observed in invasive S. pyogenes infections may be due to factors other than influenza, in part supported by the observed increase in older age groups, such as the unusually cold weather experienced in England during December 2010.The latter suggestion is supported by the observation of increases in infections in older age groups, who have been relatively unaffected by influenza.Given the on-going influenza activity in the UK, continued vigilance for changes in the incidence of S. pyogenes and S. pneumoniae infections is essential.As the start of the 2010/11 influenza season in the UK was ahead of other European countries and influenza transmission is now underway elsewhere in Europe, other national public health institutes should be alert to the possibility of similar observations.