Post-COVID STSS Surge in Japan Followed a Sharp Shift in Streptococcus pyogenes Lineages
Japan’s rebound in streptococcal toxic shock syndrome after relaxation of COVID-19 precautions was not a generic rise in invasive disease. It tracked a specific genomic replacement inside Streptococcus pyogenes, with M1UK and the newly prominent M113SNPs lineage displacing the older M1global population across 2023 and 2024.
That shift matters because STSS is among the most dangerous manifestations of group A streptococcal infection: once toxin-mediated shock and multiorgan failure begin, clinical deterioration can be rapid and unforgiving. In this nationwide laboratory-based study, 919 isolates from laboratory-confirmed cases provided a clear molecular picture of what was circulating during the resurgence.
National Surveillance Captured 919 Isolates From 1,477 Reported Cases
The analysis covered January 1, 2023 through December 31, 2024, under Japan’s notifiable disease system for STSS. Of 1,477 reported cases, 919 isolates were collected for molecular workup, giving an overall coverage rate of 62.2%; coverage was lower in 2023 at 45.9% and higher in 2024 at 68.4%. Specimens flowed through seven regional reference laboratories and the National Institute of Infectious Diseases, using standardized methods for species identification, Lancefield grouping, emm genotyping, and lineage assignment.
Among the 919 isolates, 31 emm types were identified, including two untypeable strains with signal sequence deletions. emm1 dominated the dataset at 51.1%, followed by emm12, emm89, emm49, and emm81. Together, those five types accounted for more than 85% of isolates, but the most important signal was the expanding weight of emm1 in 2024.
A second wave of replacement became visible inside emm1 itself.
M1UK and M113SNPs Replaced M1global Inside the emm1 Population
Among 470 emm1 isolates, M1UK accounted for 234 isolates, or 49.8%, and M113SNPs accounted for 176, or 37.4%. M1global fell to 60 isolates, or 12.8%. Compared with 2019, when M1global made up more than 60% of emm1 isolates, the 2023–2024 profile showed a clear lineage turnover: M1UK rose from 44% to 51%, M113SNPs from 29% to 39%, and M1global dropped from 26% to 11%.
The temporal pattern lined up with Japan’s relaxation of mask policy on March 13, 2023. STSS associated with emm1 had declined during 2020–2022, then climbed after the change in nonpharmaceutical measures, reaching 60–80 cases per month during January through June 2024. Most other types, including emm89, emm81, and emm49, remained relatively stable. emm12 also increased, but the dominant story was the resurgence of emm1-driven invasive disease.
This is the kind of epidemiologic change that usually reflects a transmission advantage, not simply a change in testing. Reduced barriers in schools, workplaces, and community settings likely gave these lineages more opportunities to circulate.
Toxin Gene Patterns Varied by Lineage and Likely Shaped Clinical Behavior
The toxin profile inside emm1 was not uniform. Of the 470 emm1 isolates, 458 were speA positive. Among those, 60.2% carried speA alone, 37.2% carried both speA and speC, and 1.7% carried speC alone; four isolates were positive for neither toxin gene.
Lineage-specific patterns were striking. M1UK mainly carried speA only, while M113SNPs usually carried both speA and speC, and M1global was also predominantly speA only. Those differences offer a plausible biological explanation for why closely related lineages may not behave identically at the bedside. Superantigen profiles can influence immune activation, cytokine storm intensity, and the speed at which shock and organ failure develop.
The lineage label alone no longer tells the whole story.
M113SNPs Was Linked to Younger Age and Higher Unadjusted Mortality
Clinical comparisons showed a concerning signal for M113SNPs. Patients with M113SNPs infection were younger, with an average age of 58.3 years, compared with 62.7 years for non-emm1 cases during the same period. Mortality was also higher: 31.3% for M113SNPs versus 23.2% for non-emm1 cases.
M1UK also skewed younger, with an average age of 54.9 years, but its mortality rate was 20.5%, similar to M1global at 20.0%. By contrast, M113SNPs stood apart. In age-adjusted comparisons, mortality remained significantly higher in M113SNPs infections than in non-emm1 infections, with an odds ratio of 1.51 and a 95% confidence interval of 1.02–2.22.
Age-stratified analysis sharpened the signal further. Mortality was significantly higher with M113SNPs among patients 25–39 years old and among those 70–84 years old. The younger-adult result is especially unsettling, because STSS often carries the greatest therapeutic urgency precisely when severe disease appears in patients who are otherwise expected to tolerate infection better than frail older adults.
Why Lineage-Level Surveillance Now Has Direct Bedside Value
The public health implication is straightforward: emm typing alone is not enough when a single emm type contains lineages with different toxin profiles and different apparent clinical behavior. M1UK and M113SNPs both expanded after COVID-era restrictions eased, but the clinical associations were not interchangeable. Surveillance that stops at emm1 would miss the finer structure that may help explain changes in severity and mortality.
For clinicians, the relevance is practical. A rising burden of invasive group A streptococcal disease means more opportunities for fulminant shock, more need for early recognition of soft tissue or primary bacteremia, and more pressure on systems already managing seasonal respiratory infections and post-pandemic circulation shifts. For laboratorians and public health teams, the message is even sharper: genomic monitoring has moved from academic detail to frontline epidemiology.
Coverage Gaps and Unmeasured Confounding Limit How Far the Mortality Signal Can Be Pushed
The study’s limitations are substantial enough to keep interpretation careful. Coverage was incomplete, at 62% overall, and lower in 2023 than in 2024. The analysis was restricted to STSS cases, so it does not describe the broader invasive streptococcal burden outside that syndrome. Clinical data were incomplete, which forced reliance on unadjusted analyses for some comparisons.
That matters because the higher mortality observed with M113SNPs could still reflect differences in host comorbidity, severity at presentation, antibiotic timing, access to critical care, or other unmeasured factors. The study therefore supports an association, not an independent causal effect. Even so, the combination of rapid lineage replacement, distinct toxin gene patterns, and a younger age distribution makes M113SNPs difficult to ignore.
Japan’s national surveillance has now documented a post-COVID expansion of emm1 STSS driven by M1UK and M113SNPs, with M1global receding into the background. Continued molecular tracking will be essential to determine whether these lineages remain dominant, diversify further, or seed similar outbreaks beyond Japan.