Background: Children with special healthcare needs (CSHCNs) face persistent barriers to timely immunization in China, but comparative evidence across disease groups and vaccines, and data on immune function, are limited. Methods: We conducted a retrospective cohort study linking electronic medical records, vaccination records, and a structured telephone and questionnaire follow-up. We estimated timely vaccination by National Immunization Program (NIP) dose definitions, assessed catch-up completion at follow-up, and compared cellular/humoral/complement immune indices with published pediatric reference ranges. Group differences used ANOVA/Kruskal–Wallis and chi-square (χ²)/Fisher’s exact tests with Bonferroni correction. Results: Timely vaccination was lower than the national healthy child benchmarks for all NIP vaccines (all p < 0.001); the Japanese encephalitis virus (JE; 24.0%) and measles-containing vaccine (MCV; 25.9%) had the lowest timely completion. A subset of CSHCNs did not receive recommended catch-up vaccinations, primarily due to persistent caregivers’ concern and point of vaccination (POV) staff’s hesitancy. Delays clustered in neonatal/perinatal disorders for Bacillus Calmette–Guérin (BCG) and hepatitis B vaccine, dose 1 (HepB1). Catch-up completion was highest for hepatitis B vaccine, dose 3 (HepB3) (86.3%) and BCG (81.8%), and lowest for the diphtheria and tetanus vaccine (DT) (49.4%); MCV2 completion was particularly low in hematological diseases. Immunoglobulin A (IgA) and immunoglobulin G (IgG) concentrations were significantly lower in neonatal/perinatal disorders and infectious disease groups versus neurological and immune disorder groups (p < 0.05). No severe adverse events were reported after catch-up. Conclusions: CSHCNs in China face substantial barriers to timely NIP immunization. Timeliness and catch-up vary substantially by vaccine and underlying condition; neonatal/perinatal disorders contribute disproportionately to early-life delays. Disease-specific guidance, strengthened POV–specialist clinic coordination, immunological monitoring, and supportive policies could improve the vaccination coverage and effectiveness in this vulnerable population.

Background: Chemotherapy, a cornerstone treatment for Acute Lymphoblastic Leukemia (ALL), can compromise immune function, leading to impaired immune memory function and diminished responses to revaccination. This systematic review and meta-analysis sought to evaluate the impact of chemotherapy on the immunogenicity of prior vaccinations and subsequent revaccination responses in children with ALL. Methods: A comprehensive search was conducted through PubMed, Embase, Web of Science, and Medline. Search time was 9 January 2025. R 4.4.2 was employed for data analysis. Results: A total of 29 relevant studies were identified, with 8 undergoing meta-analysis. The pooled antibody seropositive rates (SPR) for vaccines against Hepatitis B Virus (HBV), Hepatitis A Virus (HAV), diphtheria, tetanus, pertussis, measles, mumps, rubella, varicella, and Pneumococcal Conjugate Vaccine (PCV) demonstrated a statistically significant decline after chemotherapy in ALL patients (p < 0.0001). Subgroup analysis further revealed marked and heterogeneous declines in SPR after chemotherapy, with the magnitude of reduction varying significantly across vaccines—tetanus, HBV, HAV, measles, mumps, and rubella (Subgroup differences, p = 0.0037). Conclusions: This review provides an updated assessment of this critical topic, representing the first meta-analysis specifically focused on the effects of chemotherapy on different vaccines’ immunogenicity in children with ALL.

Purpose Summarize the characteristics of the largest cohort of BCG disease and compare differences in clinical characteristics and outcomes among different genotypes and between primary immunodeficiency disease (PID) and non-PID patients. Methods We collected information on patients with BCG disease in our center from January 2015 to December 2020 and divided them into four groups: chronic granulomatous disease (CGD), Mendelian susceptibility to mycobacterial disease (MSMD), severe combined immunodeficiency disease (SCID) and unspecified pathogenic group. Results A total of 134 patients were reviewed, and most of them had PID. A total of 112 (83.6%) patients had 19 different types of pathogenic gene mutations, most of whom (91.1%) were classified with CGD, MSMD and SCID. CYBB was the most common gene mutation (53/112). BCG disease behaves differently in individuals with different PIDs. Significant differences in sex (P < 0.001), age at diagnosis (P = 0.019), frequency of recurrent fever (P = 0.003) and infection severity (P = 0.038) were noted among the four groups. The CGD group had the highest rate of males and the oldest age at diagnosis. The MSMD group had the highest probability of disseminated infection (46.4%). The course of anti-tuberculosis treatment and the survival time between PID and non-PID patients were similar. Conclusion Greater than 80% of BCG patients have PID; accordingly, gene sequencing should be performed in patients with BCG disease for early diagnosis. BCG disease behaves differently in patients with different types of PID. Non-PID patients had similar outcomes to PID patients, which hints that they may have pathogenic gene mutations that need to be discovered.

Background/Objectives: Individual responses to recombinant human growth hormone (rhGH) therapy in children with idiopathic short stature (ISS) vary substantially, limiting pretreatment decision-making. This study aimed to develop and externally validate an interpretable machine learning model for predicting suboptimal 12-month growth response to rhGH therapy. Methods: In this retrospective dual-center study, 901 children from Center 1 were used for model development and internal testing, and 51 children from Center 2 formed an independent external validation cohort. Routinely collected baseline demographic, laboratory, hormonal, radiographic, and family-history variables were used to develop multiple machine learning models. A soft-voting ensemble classifier was constructed and interpreted using SHapley Additive exPlanations (SHAP). The primary outcome was suboptimal growth response, defined as failure to achieve a height gain of at least 0.5 standard deviation score after 12 months of treatment. Results: The optimized ensemble model showed strong discrimination in the internal test set, with an area under the receiver operating characteristic curve of 0.927, and maintained robust performance in the external validation cohort, with an AUC of 0.897. SHAP analysis identified luteinizing hormone, body mass index, TW3 RUS bone age, and insulin-like growth factor 1 as the leading contributors to predicted suboptimal-response risk. Conclusions: An interpretable ensemble machine learning model based on routinely available pretreatment data can predict suboptimal short-term rhGH response in children with ISS and may support individualized risk stratification in pediatric endocrine practice. Clinical trial registration was not required because this was a retrospective analysis.

Background: Language disorders and dyslexia are associated with atypical language- and reading-related processing, but findings from functional near-infrared spectroscopy (fNIRS) studies remain heterogeneous. This systematic review and meta-analysis synthesized existing fNIRS evidence, quantified hemodynamic differences relative to controls, and examined whether different outcomes showed different levels of consistency. Methods: PubMed, Embase, and Web of Science were searched through 9 January 2026. Eligible studies included individuals with language disorders or dyslexia assessed using fNIRS, with outcomes reported as oxygenated hemoglobin (HbO), deoxygenated hemoglobin (HbR), β estimates, or other hemoglobin-based measures. Random-effects meta-analyses were performed when sufficient quantitative data were available, with subgroup and sensitivity analyses conducted where feasible. Results: A total of 830 records were identified, of which 10 publications comprising 11 independent samples met the inclusion criteria. HbO was the only outcome to show a statistically significant overall between-group difference, with lower HbO in the experimental groups than in controls (SMD = −0.61, 95% CI [−1.12, −0.11], p = 0.017), although heterogeneity was substantial (I² = 58.6%). In contrast, pooled results for HbR and β were not statistically significant and showed weaker stability across analyses. Subgroup and sensitivity analyses did not provide sufficient evidence to support disorder category, region of interest (ROI), or outcome metric type as stable explanations for effect size variation. Conclusions: Among the included fNIRS measures, HbO showed comparatively more consistent cross-study evidence than HbR or β. However, because the HbO estimate pooled clinically distinct dyslexia and acquired language-disorder studies within a limited evidence base, it should be interpreted as a preliminary cross-diagnostic signal rather than as evidence of a shared disease continuum or common underlying mechanism.