Increased incidence of invasive group A streptococcal disease in Ireland, 2012 to 2013

M Meehan (meehanmary@yahoo.co.uk)1, S Murchan2, S Bergin3, D O’Flanagan2, R Cunney1,2,3 1. Epidemiology and Molecular Biology Unit and, Irish Meningococcal and Meningitis Reference Laboratory, Temple Street Children’s University Hospital, Dublin, Ireland 2. Health Protection Surveillance Centre, Dublin, Ireland 3. Department of Medical Microbiology, Temple Street Children’s University Hospital, Dublin, Ireland


Background
S. pyogenes is a major human pathogen that causes a wide spectrum of clinical manifestations.These range from common superficial skin infections and pharyngitis to invasive infections such as bacteraemia, meningitis, cellulitis, pneumonia, and the more severe necrotising fasciitis (NF) and streptococcal toxic shock syndrome (STSS).Invasive GAS (iGAS) infections, though relatively uncommon compared to highly prevalent non-invasive GAS infections, remains a significant global cause of morbidity and mortality.Seven-day case fatality rates range from 8% to16%, with high rates approaching 50% often associated with STSS [1].
The M protein, encoded by the emm gene, is the major virulence factor and an important epidemiological typing tool.In excess of 200 emm types have been documented to date [2].Whereas all types may be associated with invasive GAS (iGAS), emm1 and emm3 have been particularly associated with severe iGAS infections and the resurgence of these infections that has occurred since the 1980s [3].The incidence of iGAS disease exhibits seasonal patterns in addition to periodic upsurges [4].Moreover, the prevalence of emm types has been shown to exhibit temporal and geographical variations [5,6].Several factors are thought to play a role in disease fluctuations, including prominent circulating emm types, increased host susceptibility, environmental factors and seasonal viral infections such as influenza [4][5][6][7][8][9].iGAS has been notifiable in Ireland since 2004 under the Infectious Diseases Regulations 1981 as amended by S.I. no 707 of 2003.Cases are classified as confirmed, probable or possible, based on nationallyagreed case definitions, combining laboratory and clinical criteria [10].Since 2005, enhanced surveillance, which includes collection of information on isolate site, clinical presentation, risk factors and patient outcome, has been conducted for iGAS cases on a voluntary basis [11].In 2012 a laboratory reference service commenced to which isolates are voluntarily submitted for emm sequence typing.

Clinical data collection
Demographic information on iGAS cases was analysed using data which had been entered in the national electronic infectious disease reporting system (Computerised Infectious Diseases Reporting System, CIDR).CIDR allows exchange of information between laboratories, regional departments of public health and the Health Protection Surveillance Centre (HPSC) and is used to notify iGAS and record enhanced surveillance findings.

Epidemiological typing
From early 2012 onwards, all available iGAS isolates (n=109 in 2012; n=67 in January to June 2013) were submitted by clinical microbiology laboratories to the GAS reference laboratory for emm sequence typing.Isolates which had been archived in 2011 (n=28) were sent to be typed retrospectively.Typing was performed by sequencing the hypervariable region of the emm gene as previously described [12].

Statistics
Incidence rates were calculated using the 2011 population census data.Incidence rate ratios were calculated along with their 95% confidence intervals (CI; Fisher's exact test).The year 2005 was excluded from the analysis of enhanced surveillance data as there may not have been complete data reporting for that year.Difference in the emm type distribution was calculated using Fisher's exact test.

Epidemiology
In 2012, 117 confirmed cases and five probable cases of iGAS infection were notified.This corresponds to an annual incidence of 2.66 cases per 100,000 (95% CI: 2.21-3.17per 100,000).This is the highest annual incidence since iGAS became notifiable in 2004.Annual incidences for 2004 to 2011 have ranged from 0.8 per 100,000 in 2004 to 1.65 per 100,000 in 2008, with incidence rates levelling out in 2009 to 2011 (range: 1.31-1.48cases per 100,000) [11].In 2012, the highest number of cases occurred in the months May through July, with the lowest in September (Figure 1).A further increase occurred during the first six months of 2013 with a total of 102 cases notified from January to June compared to 78 and 44 cases in the first six months of 2012 and 2011, respectively (Figure 1).The number of iGAS cases in May 2013 (n=28) represents the highest number of notified cases in any month since iGAS became notifiable in 2004.
In 2012, the age-specific incidence of invasive infection increased across most age groups when compared with the average age-specific incidence over the period 2006 to 2011 (Figure 2).The biggest increases and the highest number of cases occurred in older adults (older than 75 years) followed by children (under the age of four years; p=0.053 and p=0.036, respectively) (Figure 2).Males and females were represented equally in most age groups with the exception of those in the age range 0-9 years where boys were more affected (8 female:18 male) and those in the age range 30-44 years where women were more affected (21 female:7 male).Overall, slightly more females than males were affected in 2012 (63 compared to 59); however, this was not statistically significant.
There were 26 cases of STSS in 2012 compared to an average of 6.17 cases per year for the period 2006 to 2011 (range: 3-9 cases per year).The proportion of all iGAS cases that presented with STSS increased from 10% (37 of 383) for the period 2006 to 2011 to 21% in 2012 (p=0.04).During the first half of 2013, there were 22 cases of STSS accounting for 22% of all iGAS cases, indicating that the increased severity of iGAS infections observed in 2012 is sustained thus far in 2013.
In 2012, eight patients died (12.5% case fatality rate (CFR)) within seven days of onset of the disease for which GAS was identified as the main or contributing cause of death.Of these, six patients presented with STSS.There were five deaths (11.6% CFR) in 2011, of which one case was associated with STSS, and an average of approximately three deaths per year (11% CFR) for the period 2006 to 2011, with approximately one case per year associated with STSS.For the first half of 2013, there have been nine deaths (15.3% CFR) with five cases associated with STSS.

Typing results
A total of 109 isolates (89%) collected from reported iGAS cases in 2012 were typed by emm sequence typing (Figure 3).The most prominent types were emm1, which accounted for 48.6% of isolates, followed by emm12 (9.2%) and emm28 (7.3%).Whereas, emm1 was equally distributed between males and females, emm12 and emm28 were more predominant in males (65%) and females (87.5%), respectively, although neither value was significant (p=0.34 and 0.065, respectively).The emm1 type accounted for 17 of 25 STSS cases that were typed and for seven of eight deaths.Only two isolates of emm1 were detected from October to December 2012, when no cases of STSS were reported (Figure 3).Sixty-seven isolates (66%) have so far been typed from all iGAS cases reported in January to June 2013.The number of emm1 isolates has declined somewhat, accounting for 40% of typed isolates compared to 54% of typed isolates for the same period in 2012 (January-June).However, there has been a notable increase in emm3 in the first half of 2013 with 22% (n=15) of typed isolates belonging to this emm type contrasting with 1.5% (n=1) for the same period in 2012 (p=0.0001) and 4% (n=4) in total for 2012.There is limited information regarding the type distribution of iGAS in Ireland prior to 2012.However, 28 iGAS isolates (42% of total) from 2011 were retrospectively typed and showed a lower prevalence of emm1 (25%; eight of 28) and a higher prevalence of emm12 (32%; nine of 28) compared to 2012 and 2013.

Discussion
A marked increase in notifications of iGAS occurred in Ireland in 2012 with a further increase observed in 2013.Increased incidences of iGAS have also been reported in 2012/13 in Finland, Norway, Sweden, and England, suggesting an increase in iGAS infections in Northern Europe [13,14].Public Health England reported 1,038 iGAS cases for weeks 37/2012 to 16/2013 compared to 691 for 2011/12 and 907 for 2008/09, the last peak season for iGAS infection [14], indicating that the upsurge in Ireland (April-July 2012) pre-dated that in neighbouring England.iGAS has been notifiable in Ireland since 2004.It is therefore unlikely that the initiation of iGAS typing in 2012 and alerts issued on iGAS in July 2012 and April 2013 would have had a significant effect on the overall notification rate.There is evidence to suggest that viral infections such as influenza can contribute to an increased risk of invasive bacterial infections such as iGAS [8,9].The short and mild 2011/12 influenza season in Ireland is unlikely to have had an impact on the increased incidence of iGAS in 2012.However, in 2013, the influenza season was more prolonged and may have been a contributing factor to increased rates of iGAS [15].
The high prevalence of emm1 is likely to have played a significant role in the increased incidence of iGAS in 2012.The lower prevalence of emm1 and higher prevalence of emm12 in 2011 compared to the incidence of these types in 2012 suggests a possible change in the epidemiology of GAS between 2011 and 2012.However, no major conclusions can be drawn due to the small number of isolates which were typed in 2011.Significantly, England, Norway and Sweden have also reported a high prevalence of emm1 in 2012/13 [13,14].The increased incidence of emm3 in 2013 is of concern given its association with a higher case fatality rate compared to other emm types [16].Significantly, in 2013, emm3 accounted for five of eight fatal cases in Ireland for which typing data was available.An increased prevalence (11%) of emm3 was also reported in Sweden in 2012 [13].Continued typing of isolates will be required to confirm this trend in 2013 and to monitor any further changes in epidemiology.
The incidence of iGAS infections in Ireland, although increased in 2012, was lower than the 2012 upsurge rate reported in Sweden (6.1 per 100,000 population) [13].Moreover, the overall annual incidence rates (0.8-1.65 in  in Ireland prior to the upsurge were lower than that reported in the United Kingdom (UK), the Nordic countries and North America, which have in general higher reported incidence rates in the range 2.5-5.0 per 100,000 [4,11,13].The reason(s) for the lower incidence rate of iGAS in Ireland is unclear [11].A relatively high empiric antibiotic consumption has previously been suggested as a likely explanation [11].For example, in 2010, outpatient penicillin use was higher in Ireland at 10.70 defined daily doses (DDD) per 1,000 inhabitants per day than in Finland at 6.6 DDD, Sweden at 7.1 DDD and the UK at 8.6 DDD per 1,000 inhabitants per day [17].A small number of cases may also have gone unreported due to differences in case definitions between Ireland and other countries as previously reviewed [11].Temporal and geographical fluctuations in iGAS infections may also be due in part to emm type distribution, population susceptibility, prevalence of certain risk factors, and environmental factors such as climate and population density [4,11,18].
Similar to other European countries, seasonal variation in iGAS infections was observed in Ireland, with lowest incidence rates in late summer/autumn as reported elsewhere [4,13,19].The peak rate of iGAS infections in Ireland in May 2012 and 2013 occurred later than in Sweden and other Nordic countries which generally exhibit peak incidences in the first quarter of each year [13,19].However, occasional peaks in summer can also occur in these countries [13,19].Noteworthy is the rising incidence of iGAS infections reported in April 2013 in England, possibly indicating a late upsurge similar to that observed in Ireland [14].Seasonal variations in iGAS may reflect climatic differences such as daylight/ sunlight hours, rainfall/humidity, temperatures and seasonal behavioural patterns [4,19].Certainly, spring 2012 and 2013 exhibited below average temperatures throughout Ireland.
Given the severity and rapid progression of iGAS, prompt detection and medical intervention are the only preventative and control measures available to reduce morbidity and mortality.In response to the increase in iGAS notifications, alerts and guidance have been issued to public health departments and clinicians (particularly general practitioners, emergency physicians, paediatricians, infection specialists and intensive care specialists), and iGAS recognition and management have been highlighted in surveillance bulletins.Laboratories are encouraged to send iGAS isolates for typing and to submit enhanced surveillance data to monitor the continuing trends in iGAS epidemiology, to investigate any suspected outbreaks and improve our understanding of iGAS epidemiology in Ireland.Moreover, in light of the reported increase in iGAS in several other European countries and the changing epidemiology of iGAS detected here in Ireland, public health institutions in Europe should be alert to any changes in epidemiology for the forthcoming 2013/14 iGAS season.Consideration should also be given to expanding the European Invasive Bacterial Disease Surveillance Network (EU-IBD) to include iGAS.

Figure 1
Figure 1Invasive group A streptococcal cases notified by month and year in Ireland, January 2006-June 2013 (n=607)

Figure 2
Figure 2Age-specific incidence rate of invasive group A streptococcal infections per 100,000 population by five-year age groups, Ireland, 2012 and averaged over the period 2006-11