This five-year retrospective study examined invasive bacterial infection in a normally sterile site in the head, neck, chest, blood and CSF. There was a much higher rate of invasive bacterial disease than in the four preceding winters (including pre-pandemic years) and a statistically significantly higher case fatality of 9%. Children who died in 2022 had no evidence of abscess formation from their infection. Only one fatality had a pre-existing neurodisability, suggesting the documented infections were overwhelming usually healthy young children.
A surge in invasive GAS was noted in Europe, including increased rates of GAS empyemas over the same time period in 2022 (14, 15). Locally in Ireland, the national HPSC alerted physicians to the prevalence of GAS disease in the community (16, 17). A pre-pandemic multicentre European study published data from 2012–2016 re-outlined that invasive GAS infection is associated with significant morbidity and mortality in the paediatric cohort, and that it can be a life-changing disease for previously well children (18), with a 2% case fatality (18), similar to the numbers we found in the pandemic years, but much lower than 2022 (9%). While GAS was the predominant bacterial pathogen identified in our group, there was also increased detection of other pathogenic bacteria associated with abscesses in normally sterile head, neck and chest sites in 2022. Our study is comparable with emerging international literature looking at children with sepsis, quoting mortality rates of around 9% (19, 20). To date, no vaccine is licenced against GAS for use in the community. The other three notifiable diseases we studied are, at least partially, vaccine-preventable (21). Invasive GAS disease in children is often linked to a preceding varicella infection (22, 23), but we did not see this in our patient group. Varicella immunisation has only been recommended to be included in Ireland’s primary immunisation schedule in July 2023 (24, 25). The influenza vaccine was introduced as a free vaccine in Ireland for children in 2020 but with subsequent low levels of uptake (26). This trend continued in 2022 with only 15.4% of children < 17 years of age receiving a vaccine (27).
The surge in invasive bacterial diseases could be directly linked to the behavioural and societal changes during the Covid-19 pandemic, where children were naive to infections due to social distancing, school closures and isolating in family units (28, 29). With the gradual reopening of society in 2021, it was noted worldwide in America, Australia and Europe that there were significant outbreaks of RSV infection compared to low rates during the pandemic (30–32). There were also low rates of antibiotic use compared to pre-pandemic numbers (33) and reduced hospital admission rates (34). Reduced exposure to bacterial and viral pathogens in the community setting has potentially led to an “immunity debt” with decreases in both innate immunity and adaptive/specific immunity (29, 35). There was a concern that this could lead to a surge of infections not normally seen in the community setting (28). Invasive bacterial diseases, particularly those caused by bacteria normally carried in the upper respiratory tract, such as S. pneumoniae, N. meningitidis, and H. influenzae, significantly declined during the pandemic (36–38). It now appears that there was a surge of invasive bacterial diseases in Irish children after the lockdown. Staph Aureus, which is normally one of the “classic” organisms associated with invasive bacterial disease (39, 40) represented a small number of our cohort, which may reflect the greater role of organisms associated with respiratory carriage in this data set, along with excluding bone or joint infections.
Many children had presented to a healthcare facility before the presentation leading to admission, and while this is not reported in recent studies looking at invasive infections in children, there is a noted lag in times leading to admission in both PICU and non PICU patients (18). GAS can colonize the pharynx of asymptomatic individuals (41) and is easily treatable but in some children, it can lead to invasive and significant infection (14, 42). Our study affirms a recent study in Spain that demonstrated a surge in hospital presentation and ICU admission in late 2022 compared to pandemic and pre pandemic years (43).
The strengths of our study were rigorous criteria for case definition and regular multi professional review of cases between general paediatrics, paediatric emergency medicine, microbiology and radiology. Case ascertainment was maximised by use of thorough complimentary and overlapping searches across diagnostic disciplines and data sets using the same criteria in all three sites of this study over all five years of the study.
Possible weaknesses of our study are that despite best efforts, cases may have been missed, especially if they were transferred into CHI from outside CHI and had all their imaging and investigations carried out in their local centre. We countered this anticipated issue by searching both radiology and microbiology databases. The Irish Meningitis and Sepsis Reference Laboratory (IMSRL) is based at CHI Temple Street and provides a national PCR diagnostic service for invasive bacteria, so results of PCR diagnostics carried out prior to transfer were available at CHI. Other cases nationally were not referred to CHI and were managed locally, so this is not a national data set. However, as CHI has the only PICU capacity in the Republic of Ireland it is likely that all the most severe and fatal cases are included in this cohort. We had a strict case definition which meant that all included cases are “true” cases and we have not included cases that went outside of our case definition, therefore the burden of illness related to these pathogens may not be fully included in the case definition. As CHI does not have a full electronic health record, the study required personal review of paper medical charts. Given the retrospective nature of this study, data collection was limited to what was recorded in real-time in the paper and electronic medical records. We attempted to collect influenza vaccination rate, COVID-19 infection, COVID-19 vaccination status and parental smoking status. Parental smoking, in particular, is linked to increased rates of invasive bacterial disease (44). All of these were poorly documented in the admission and inpatient medical notes in contrast to documentation of infant immunisations, which was almost always complete. Having incomplete or absent primary childhood vaccination details can lead to additional investigations on presentation to a paediatric ED (45). Clinicians and nursing staff may be more primed to ask about primary childhood vaccines, while influenza and COVID-19 vaccine status could have seemed incidental as they were currently additional or optional vaccine programmes at the time.