Pediatric Mycoplasma pneumoniae activity increased sharply beginning in mid-2023 and culminated in a large resurgence by 2024, according to the multisite CGMH surveillance reported in this study. Case incidence in 2024 doubled relative to the 2019 peak, reaching 546 per 100,000 population. Admissions to intensive care units increased 2.2-fold to 10.9 per 100,000, and central nervous system (CNS) complications rose 13.7-fold to 6.1 per 100,000. The authors reported that this resurgence was unusually severe in Taiwan compared with some other countries, and they framed the trend against known postpandemic respiratory pathogen behavior—slower reemergence of M. pneumoniae than of viruses such as influenza or RSV because of the organism’s long incubation and growth characteristics.
Investigators used the Chang Gung Memorial Hospital (CGMH) network laboratory surveillance system across four branches (Linkou, Taipei, Tucheng, and Kaohsiung) to identify pediatric cases from January 2017 through December 2024. Diagnostic testing was clinician-driven for children with compatible clinical syndromes (acute lower respiratory infection, persistent fever, or suspected CNS manifestations). Respiratory specimens were tested by real-time PCR; beginning in 2019, laboratories also used the BioFire FilmArray Respiratory Panel. For suspected CNS involvement, paired CSF and respiratory specimens underwent PCR testing. Population denominators were estimated using Taiwan’s National Health Insurance data to derive annual and monthly incidence rates for hospitalizations, ICU admissions, and M. pneumoniae–related CNS complications.
The authors extended a prospective cohort of children (<18 years) with laboratory-confirmed M. pneumoniae infection during 2023–2024. They excluded patients with chronic lung disease, genetic or metabolic disorders, neurologic disease, immunodeficiency, or cancer. In the cohort (n = 226), severe infection was defined as admission to an ICU. Multivariable analysis identified four independent risk factors for ICU admission: neutrophil count >7,000/µL (adjusted odds ratio [aOR] 6.2), platelet count >400 × 10^3/µL (aOR 12.5), presence of pleural effusion (aOR 9.7), and CNS involvement (aOR 27.9). The study team reported those aORs with 95% confidence intervals in the source. The report did not supply in-text details of other treatment decisions, such as which second-line antimicrobials clinicians used when macrolide resistance was suspected.
Investigators performed multilocus sequence typing (MLST) and 23S rRNA PCR on isolates and analyzed a phylogeny of 698 genomes. MLST used the PubMLST scheme with eight loci to assign sequence types. Phylogenetic analysis demonstrated that four novel subclades of sequence type 3 (ST3) emerged after the pandemic and had replaced pre–COVID-19 lineages by mid-2024. One of these emerging subclades was associated with severe clinical outcomes in the analyzed dataset. The authors noted Taiwan’s historically high macrolide-resistant rates—about 70% prior to the pandemic—and framed the genomic observations in that epidemiologic context. The source text did not include full MIC distributions or comprehensive per-subclade resistance percentages in the body provided here; those details may appear in the cited tables and appendices.
The study defined M. pneumoniae–related CNS complications as neurologic syndromes temporally associated with infection and confirmed by PCR detection of M. pneumoniae in respiratory or CSF samples when no alternative etiology was identified. Included manifestations were encephalitis, brainstem or basal ganglia encephalitis, acute disseminated encephalomyelitis, cerebellitis, Guillain–Barré syndrome, transverse myelitis, and isolated encephalopathy. For children with CNS involvement, paired CSF and serum were collected on the same day. The authors measured CSF and serum IgG and albumin concentrations by commercial ELISA kits and calculated the CSF IgG index using a standard formula that the paper reproduced. The source provided the laboratory approach and formula; it did not report, within the excerpt used here, specific numeric CSF IgG index thresholds tied to clinical outcomes across the cohort.
The authors acknowledged evolving pathogen genetics and increasing severe presentations but left several items unresolved in the text excerpt. The surveillance approach relied on clinician-ordered testing, so ascertainment could vary with practice patterns; the study’s reliance on the CGMH network covers a large population, but the authors used National Health Insurance data to estimate catchment denominators rather than direct population registers. The body of the report here did not present detailed antimicrobial management strategies for macrolide-resistant infections, full MIC distributions, or granular demographic breakdowns beyond cohort inclusion and exclusion criteria. The phylogenetic finding that one ST3 subclade associated with severe outcomes raises specific clinical questions: whether virulence determinants or resistance mutations explain the link, how that subclade spreads regionally, and what therapeutic adjustments clinicians should adopt. The source calls for enhanced surveillance and further analysis; the text included in this summary ends on those practical research and public health priorities rather than on any broad speculative conclusion.