Introduction
Streptococcus dysgalactiae subspecies
equisimilis (SDSE) was proposed as a new subspecies in 1996 [
1]. SDSE isolated from humans as a commensal microorganism possesses antigens belonging to Lancefield group G, C, or A [
2]. Although SDSE forms relatively large glossy colonies on blood agar that are strongly β-haemolytic, several biochemical property tests, such as the pyrrolidonyl arylamidase test, are required to distinguish SDSE from other haemolytic streptococci. Moreover, SDSE had been considered substantially less pathogenic than
Streptococcus pyogenes (GAS). However, whole-genome sequencing has revealed that SDSE possesses several virulence factors. These factors include M protein (
emm), streptokinase, C5a peptidase, hyaluronidase, and others involved in immune evasion, as demonstrated in GAS, and systemic toxicity of streptolysin O and streptolysin S [
3,
4]. Thus, SDSE and GAS are considered closely related phylogenetically and may have originated from a common precursor [
5].
SDSE was considered an asymptomatic coloniser of the human upper airway, skin, and gastrointestinal tract. Recently, it has been recognised as a clinically relevant pathogen that causes a broad range of diseases, from milder illnesses such as local skin and soft tissue infections to life-threatening conditions such as streptococcal toxic shock syndrome (STSS) and necrotising fasciitis (NF), similar to those caused by GAS [
6‐
12]. Invasive SDSE (iSDSE) infections are more common in older adults [
12,
13] and individuals with underlying chronic diseases [
7‐
10,
14,
15]. Consequently, the impact of disease burden associated with SDSE infections will grow in developed countries, which are aging rapidly.
Although studies have compared the onset age [
16,
17], underlying diseases [
7,
14‐
16], biomarkers [
12], and mortality rates of iSDSE [
7‐
9,
12,
14] with invasive GAS and invasive Group B streptococcal infections, only a few have clarified the risk factors for death in patients with iSDSE [
17]. A common problem exists in that clinicians have limited awareness of the pathogenesis of SDSE. Therefore, we aimed to consolidate clinically relevant information on SDSE infections. We thus conducted this large-scale surveillance study of iSDSE infection in Japanese adults to evaluate the relative contribution of host factors, clinical manifestations, biomarkers, and bacterial factors to the 28-day mortality after admission.
Discussion
In this most extensive surveillance of adults with iSDSE infections in Japan, we revealed an overall 28-day mortality rate of 10.4%, which was significantly associated with older age (≥ 60 years), disease severity, clinical manifestations, biomarkers on admission, and macrolide resistance.
As expected, disease severity on admission was an independent risk factor significantly associated with mortality. Clinical evaluation of disease severity according to the guidelines is crucial for physicians to promptly predict iSDSE infection prognosis.
Here, the number of patients and mortality rate increased with age, and advancing age had a large effect on the mortality rate. Moreover, most patients with iSDSE infections had various underlying diseases. Inflammatory responses to SDSE infection may be impaired with aging, leading to fragility and increased susceptibility to recurrent or persistent infections. Chronic inflammation associated with aging, termed ‘inflammaging’, is a risk factor for diabetes mellitus, malignancy, cardiovascular disease, and kidney disease [
23]. We hypothesise that these processes, including aging, SDSE colonisation, and underlying disease progression, contribute considerably to the morbidity and mortality of patients with SDSE infections in Japan, which is a super-aged society.
Bacteraemia without primary focus and other invasive infections (such as bacteraemic pneumonia, NF, meningitis, and STSS) had poor prognosis compared to cellulitis. The fatality rates were 18.0 and 12.4%, respectively, which are consistent with those reported in previous studies, with 13 to 20% [
7‐
9,
14,
24].
Regarding the laboratory findings, WBC < 4,000 cells/µL, Cr ≥ 2.0 mg/dL, and CK ≥ 300 IU/L were important biomarkers and correlated strongly with poor outcomes. Studies describing biomarkers as predictors of mortality in patients with SDSE are limited. One previous study reported that patients with SDSE infections had an increased mortality rate with leukopenia (WBC < 5,000 cells/µL) and thrombocytopenia (PLT < 13.0 × 10
4 cells/µL) [
13]. International guidelines [
18,
20,
21] on severe conditions in patients with each disease particularly emphasise leukopenia (WBC < 4,000 cells/µL). Additionally, acute tubular necrosis resulting from sepsis-induced tissue hypotension and/or hypoxemia may contribute to renal injury, as shown by the laboratory findings such as Cr ≥ 2.0 mg/dL. Moreover, the serum concentrations of CK, which is released into the bloodstream when muscle cells disintegrate, increased in patients with poor outcomes. Thus, high CK concentrations may reflect extensive or systemic injury caused directly by SDSE infections or indirectly by hypoperfusion, indicating multiple organ damage rather than a specific diagnosis. Our study also revealed that patients infected by isolates harbouring the macrolide-resistance gene had a significantly poor prognosis. The mechanism and clinical relevance remain to be explored.
Regarding the
emm type,
stG6792 has constantly been the most prevalent in Japan [
12,
25], and it is also emerging in other countries. In European countries,
stG62647 has recently become more prevalent [
13,
26,
27] in addition to
stG643,
stG485, stG6, and
stC74a [
10,
11,
28,
29]. According to the CDC database concerning the
emm type, the
stG6792 reference strain appears to be derived from India, suggesting that this type was transmitted from India to Japan. The routes of transmission and the reasons for variation in the dominant
emm types among countries should be elucidated.
Our study has some limitations. First, although we evaluated mortality risk according to the clinical manifestations based on three groups (cellulitis, bacteraemia without primary focus, and others), the risk of death for each disease might have been underestimated. For example, NF and STSS, the most serious forms of iSDSE infections, were included in the ‘others’ category, which includes various invasive diseases. However, we elucidated a poor prognosis of bacteraemia without a primary focus compared to cellulitis. Second, the number of patients with STSS, which is associated with mortality, was small (n = 4) in this study, and the proportion of STSS from SDSE is lower than that from GAS. We conducted a six-year surveillance program for the three streptococcal species GAS, SDSE, and
S. agalactiae (GBS) as part of the “Nationwide Invasive Streptococcal Disease Surveillance”. As a result, STSS cases in adults were 7.1% (23/326) for GAS and 0.7% (4/588) for SDSE, and 0.5% (2/443) for GBS. We included patients with invasive SDSE, which was defined as isolation of SDSE from normally sterile clinical samples such as blood, cerebrospinal fluid, and closed pus. However, according to the study protocol, we excluded patients with SDSE from a wound culture accompanied by necrotizing fasciitis or streptococcal toxic shock syndrome. In addition, the unique Japanese “universal health insurance system”, which provides good access to medical institutions and makes it easy to prescribe antibiotics could have had an impact on morbidity and mortality. Moreover, the differences in lifestyle habits between Japan and western countries may have influenced the results. Future surveillance will need to elucidate this issue through global collaboration. Third, we could not analyse missing data (lack of basic information or loss on follow-up), which accounted for 14.3% (n = 98) of the total cases based on our retrospective design. However, we confirmed that the demographics, clinical manifestations, and prognosis of those patients did not significantly differ from those included in this study. Fourth, we categorised disease severity into two groups, ‘mild-to-moderate’ and ‘severe’. As shown in Table
1, various diseases comprise iSDSE infections. For example, the practice guidelines for skin and soft tissue infection [
18] define disease severity as “mild,” “moderate,” and “severe,” whereas those for other diseases [
20,
21] categorise severity into “non-severe” and “severe.” Therefore, we divided patients with iSDSE infections into “mild-to-moderate” and “severe” groups. Finally, the proportion of antibiotic classes as the initial therapy differed between the survival and mortality groups because the empiric antibiotic treatment was influenced by apparent disease severity. Furthermore, combination therapy of penicillin with clindamycin for life-threatening infections was given to a small number of patients, and its efficacy remains unclear.
In conclusion, our results provide evidence of prognostic factors for death due to iSDSE infections and the basis for evaluating clinical manifestations, disease severity, and biomarkers on admission. The prevalence and burden of iSDSE infection, which particularly affects older adults with underlying diseases, are likely to further increase in developed countries. Strategies to reduce the mortality risk of iSDSE infections require further investigations of dynamic molecular epidemiology and clinical characteristics, as well as optimal treatment based on well-conducted studies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.