A Machine-Learning Model for the Prognostic Role of C-Reactive Protein in Myocarditis

Area of Science:

• Cardiology
• Inflammation
• Biomarkers

Background:

• The role of inflammation markers, such as C-reactive protein (CRP), in diagnosing and predicting outcomes for myocarditis remains unclear.
• Understanding these markers is crucial for effective patient management and prognosis.

Purpose of the Study:

• To evaluate the diagnostic and prognostic significance of C-reactive protein (CRP) levels at the time of diagnosis in patients with myocarditis.
• To identify predictors of mortality and heart transplant (Htx) in this patient cohort.

Main Methods:

• Retrospective analysis of 409 patients with clinically suspected (CS) or biopsy-proven (BP) myocarditis, with available CRP data at diagnosis.
• Machine learning (random forest) was used to analyze survival data and identify predictors of death/Htx.
• Clinical, laboratory, and imaging data were collected at diagnosis and follow-up.

Main Results:

• Abnormal CRP levels were more frequent in CS myocarditis, associated with recent viral infection, shorter symptom duration, chest pain, better functional class, and higher troponin I.
• There were 13 deaths/Htx events over a median follow-up of 2.9 years (10-year survival 94%), with 10 occurring in BP myocarditis patients.
• CRP levels did not significantly differ survival rates (p = 0.23).

Conclusions:

• Elevated CRP at diagnosis is associated with less severe clinical presentations in CS myocarditis but does not predict survival.
• The strongest predictors of death/Htx were reduced left ventricular ejection fraction (LVEF), biopsy-proven myocarditis, and positive anti-nuclear auto-antibodies (ANA).