Group A streptococcus (GAS) can cause a range of clinical syndromes, including superficial infections such as pharyngitis and impetigo, post-streptococcal immunological complications such as acute rheumatic fever and glomerulonephritis, and invasive infections. The direct costs of GAS-related disease in New Zealand have been estimated at almost 30 million NZD (2015 costs) per year.[[1]]

Invasive GAS (iGAS) infections occur when GAS infects a normally sterile site, such as joints, deep tissues, pleural fluid, cerebrospinal fluid or, as in this study, the bloodstream. Such infections can be severe and sometimes life-threatening and thus require early recognition and treatment. In high-income nations the incidence of iGAS typically ranges from 2 to 4 cases per 100,000.[[2]] In contrast, a much higher incidence of iGAS has been observed in some disadvantaged populations (106 cases per 100,000 in Indigenous Australian people in the Northern Territory,[[3]] 46 cases per 100,000 in Native American people in Arizona) and lower income nations (12 cases per 100,000 in Fiji, 13 cases per 100,000 in children under 16 years in Kenya).[[2]] Mean case-fatality rates from iGAS in high income countries range from 8% to 16%.[[2]]

In New Zealand, iGAS is not currently notifiable, despite being nationally-notifiable in a number of other countries including Australia, the United Kingdom (UK) and Canada.[[4–6]] Recently in the UK (2022), this notification system has led to detection of a spike in iGAS incidence and mortality, with a nationwide public health response.[[7,8]] A New Zealand study estimating the incidence of iGAS reported a significant upward trend from 3.9 per 100,000 in 2002 to 7.9 per 100,000 in 2012.[[9]] Due to the lack of notification, the more recent incidence of iGAS is uncertain. Current surveillance is primarily laboratory-based, relying on individual laboratories voluntarily sending clinically relevant GAS isolates to the Institute of Environmental Science and Research (ESR) for further typing.[[10]] Therefore, this study aimed to describe the epidemiology, demographics, clinical features and healthcare impact of iGAS disease in Hawke’s Bay over six years (2016 to 2021), to inform public health efforts.

Methods

Definitions

The case definition for iGAS disease in this study was isolation of GAS from blood culture. Severe iGAS disease was defined as cases requiring intensive care unit (ICU) admission or with death from any cause within 60 days of diagnosis.

Study design and data collection

Cases of iGAS disease were identified retrospectively from the Te Whatu Ora Te Matau a Māui Hawke’s Bay microbiology laboratory blood culture database, including specimens collected from 1 January 2016 to 31 December 2021. Further clinical information was obtained from patients’ electronic health records (Clinical Portal).

Data analysis

Population data from the 2018 New Zealand Census (the approximate midpoint of the study period) were used for incidence calculations.[[11]] Haemodialysis population data were obtained from Australia and New Zealand Dialysis and Transplant Registry (ANZDATA) Annual Reports for 2016–21, and included Hawke’s Bay Hospital, satellite and home prevalent haemodialysis patients.[[12]] Age-standardisation was calculated using the 2001 New Zealand Census total Māori population data as the standard population.[[13]] New Zealand Index of Deprivation (NZDep) 2018 data were used for assessment of socio-economic deprivation.[[14]] Residential address at diagnosis was used to epidemiologically screen for household clusters of iGAS cases. Data analysis was performed using Microsoft Excel®.

Ethics approval

Ethics approval was sought from the New Zealand Health and Disability Ethics Committees. The study was deemed out of scope and not requiring ethics review. The study was approved by the Hawke’s Bay Clinical Research Committee (2023/01/371).

Results

A total of 93 cases of iGAS disease were identified in the Hawke’s Bay Region between 2016 and 2021. The median age of cases was 64 years (interquartile range [IQR] 42–76 years). Table 1 shows that the overall age-standardised incidence of iGAS disease in Hawke’s Bay was 5.6 per 100,000 (95% confidence interval [CI] 4.1–7.4). People of Pacific ethnicity had the highest age-standardised incidence at 14.5 per 100,000, followed by Māori and Asian peoples (see Table 1). However, case numbers were small in the Asian and Pacific cohorts.

View Figures 1–4, Tables 1–2.

Approximately half of all iGAS cases occurred in people aged 65 years and over (n=50, 53%). Within the 65 to 79 year olds age group, there was an incidence of 80.4 per 100,000 in Māori (13 cases, 95% CI 42.8–137.5), compared to 10.5 per 100,000 in the European/Other ethnicity group (13 cases, 95% CI 5.6–18.0).

In children under five years of age there was an incidence of 9.4 per 100,000 population (eight cases, 95% CI 4.1–18.5), with Pacific, Asian and Māori children disproportionately affected (see Figure 1). Cases in children under five years were caused by cellulitis (six cases), pneumonia (one case) or bacteraemia of unknown source (one case).

In the 15 to 34 year olds age group, there was an incidence of 4.4 per 100,000 in females (five cases, 95% CI 1.4–10.3), compared to 0.8 per 100,000 in males (one case, 95% CI 0.0–4.6). Cases in females in this age cohort were caused by postpartum endometritis (two cases), lactational mastitis and pneumonia. There was also no statistically significant difference in incidence of iGAS disease between males and females in the under 5 or over 64 year olds age groups.

No household clusters of iGAS were epidemiologically identified, including in mother-baby pairs. There were two possible clusters in aged residential care facilities (each with two cases diagnosed in residents from the same facility within 7 days). However, no molecular typing results were accessible for laboratory confirmation.

There was a significantly higher incidence of iGAS in areas of greatest socio-economic deprivation (NZDep score 9–10; 14.9 per 100,000 population [95% CI 10.7–20.2]) than in the least deprived areas (NZDep score 1–2; 4.0 per 100,000 [95% CI 1.5–8.7]) (see Figure 2). Nine of the 11 children under 10 year olds that developed iGAS were living in NZDep score 8–10 areas (82%).

The annualised incidence of iGAS disease ranged between 3 and 9 per 100,000 over the study period (see Figure 3).

Clinical features, severity and management

Common pre-existing medical risk factors for iGAS disease included diabetes mellitus (31%), chronic kidney disease (29%) (see Table 2). One patient (included in the chronic kidney disease cohort) was receiving haemodialysis.

Skin infections were the most common source, with cellulitis diagnosed in 65 cases (70%) (see Figure 4). Of the 93 cases, 37 (41%) met criteria for severe iGAS disease. ICU admission was required for 29 patients (31%). Median length of hospital stay was 6 days (IQR 3–11) and the total combined length of hospital stay for all patients was 991 days over the 6-year study period. The 60-day all-cause mortality was 12% (11 deaths). The age range of people who died following iGAS infection, was 42 to 96 years (median 75 years, IQR 67–86). Of these deaths, eight cases were associated with skin infections (cellulitis or necrotising fasciitis), and three with bacteraemia of unknown source.

Discussion

Invasive GAS disease is an important source of morbidity and mortality in Hawke’s Bay, particularly in children under five years of age and adults over 65 years. When compared to people of European or Other ethnicities, the age-standardised incidence of iGAS disease was 7-fold higher in Pacific peoples and 5-fold higher in both Māori and Asian populations. As with other GAS-related disease, people residing in areas of high socio-economic deprivation were significantly over-represented.[[15]] The annual incidence of 5.6 cases per 100,000 population (95% CI 4.1–7.4) described in this study is similar to rates previously reported in New Zealand, but higher than findings from other high income countries such as the UK and Canada.[[1,9,16,17]] As shown in Figure 3, the incidence of iGAS disease was lower in 2020–21 than in 2019. This trend may have been influenced by the COVID-19 pandemic, with public health measures to prevent COVID-19 concurrently reducing transmission of GAS.

Skin infections are the most common source of iGAS disease in Hawke’s Bay (including in both peak age groups of under five and over 65 years), providing further evidence for the importance of action on the social determinants of skin health in prevention of GAS-related disease. Previous studies have demonstrated an association of GAS-related disease with household overcrowding and socioeconomic disadvantage.[[15,18]] The Child Healthy Housing Program (part of the national Healthy Homes Initiative), which commenced in Hawke’s Bay in 2015, provides an example of a successful prevention strategy addressing underlying social determinants of health. This program has since received additional funding and been expanded nationally.[[19,20]]

The Hawke’s Bay School Skin Program is an equity-driven primary prevention program that commenced in 2020 in response to the high local rates of serious skin infections.[[21]] Given the significant burden of disease attributable to skin-related iGAS infections in those aged under five years, work is underway to further extend this Program into the early childhood education sector, particularly focussing on children living in areas of socio-economic disadvantage.

Known predisposing factors for iGAS disease in children include varicella zoster virus infection, influenza infection, trauma, burns, surgery, immunocompromisation, malignancy and age under one year.[[22]] Scabies infection has been shown to be a contributing factor to skin infections among children in remote Indigenous Australian communities and could be further studied in New Zealand.[[23]] In our study, the most frequent comorbidities in the older age cohort with iGAS disease were diabetes mellitus and chronic kidney disease.

There were higher rates of severe iGAS disease (41%), requiring ICU admission or causing death, in this study than found in other jurisdictions that used a broader definition of severe disease.[[3,24]] Possible explanations may include a lower local case detection rate for non-severe cases, receipt of antibiotics prior to blood cultures, or delayed diagnosis and treatment. Review and optimisation of blood culture collection practices is recommended. Improving access to primary care, particularly for high-risk population groups, may help to reduce disease severity through earlier detection and treatment of GAS infections.

This audit will have underestimated the true burden of iGAS disease in Hawke’s Bay, as cases were only identified if GAS was isolated from blood culture and not from other sterile sites. Other sterile sites were not included as these data were less accessible and would require detailed review and clinical judgement against a broader case definition. However, research from regions with active iGAS surveillance (including non-blood sterile sites) have found blood cultures to be positive in 94% of cases.[[3]] Cases may also have been missed if blood cultures were collected after antibiotic exposure or less than the recommended volume of blood was collected.

Small case numbers in some sub-groups is an important limitation of the incidence data. The high incidence of iGAS in people of Asian ethnicity (10.5 per 100,000) is driven by only six cases, with a wide 95% confidence interval (3.4–23.7). The high incidence of iGAS in people of Pacific ethnicity was less apparent in crude (Figure 1) than in age-standardised results (Table 1) because of the comparatively young age distribution of the Pacific population in Hawke’s Bay.[[11]]

Further action is required to address the underlying social determinants of skin health and inequities in access to primary care in Hawke’s Bay, particularly in the highest-risk population groups.[[15,18]] Mandatory national notification of iGAS disease would provide the opportunity for improved surveillance of GAS-related disease, and consideration of a public health response to iGAS disease in New Zealand.