Objective:

To investigate the clinical characteristics and cytokine profiles of children with Mycoplasma pneumoniae pneumonia (MPP) and to establish a predictive nomogram for the early recognition of severe Mycoplasma pneumoniae pneumonia (SMPP).

Approach:

• Study Design: A prospective study enrolling 445 children with MPP, categorized into mild MPP (MMPP, n = 190) and SMPP (n = 255) groups based on disease severity.
• Data Collection: Clinical features and laboratory parameters were compared between groups. Binary logistic regression analysis was used to identify independent risk factors for SMPP.
• Cytokine Analysis: Serum cytokine levels were measured in a sub-cohort of 84 children selected from both groups using stratified random sampling.

Key Findings:

• 57.3% of children developed SMPP.
• Independent risk factors for SMPP included wheezing (OR=4.016), shortness of breath (OR=4.717), D-dimer (OR=3.032), CRP (OR=1.033), fever (OR=3.432), and age (OR=1.114).
• The predictive nomogram model had an AUC of 0.818, indicating good predictive ability.
• Serum levels of HMGB-1, TFEB, MCP-1, MCP-2, MCP-3, MCP-4, and TNF-α were significantly higher in the SMPP group.

Interpretation:

A predictive nomogram incorporating key clinical features and laboratory parameters can aid in early risk stratification of SMPP in hospitalized children.

Limitations:

• The study was conducted at a single center, which may limit generalizability.
• The sample size for cytokine analysis was relatively small.

Conclusion:

The study established a predictive nomogram for early identification of SMPP.