Back to Search
Start Over
A Machine-Learning Model for the Prognostic Role of C-Reactive Protein in Myocarditis.
- Source :
-
Journal of Clinical Medicine . Dec2022, Vol. 11 Issue 23, p7068. 13p. - Publication Year :
- 2022
-
Abstract
- Aims: The role of inflammation markers in myocarditis is unclear. We assessed the diagnostic and prognostic correlates of C-reactive protein (CRP) at diagnosis in patients with myocarditis. Methods and results: We retrospectively enrolled patients with clinically suspected (CS) or biopsy-proven (BP) myocarditis, with available CRP at diagnosis. Clinical, laboratory and imaging data were collected at diagnosis and at follow-up visits. To evaluate predictors of death/heart transplant (Htx), a machine-learning approach based on random forest for survival data was employed. We included 409 patients (74% males, aged 37 ± 15, median follow-up 2.9 years). Abnormal CRP was reported in 288 patients, mainly with CS myocarditis (p < 0.001), recent viral infection, shorter symptoms duration (p = 0.001), chest pain (p < 0.001), better functional class at diagnosis (p = 0.018) and higher troponin I values (p < 0.001). Death/Htx was reported in 13 patients, of whom 10 had BP myocarditis (overall 10-year survival 94%). Survival rates did not differ according to CRP levels (p = 0.23). The strongest survival predictor was LVEF, followed by anti-nuclear auto-antibodies (ANA) and BP status. Conclusions: Raised CRP at diagnosis identifies patients with CS myocarditis and less severe clinical features, but does not contribute to predicting survival. Main death/Htx predictors are reduced LVEF, BP diagnosis and positive ANA. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 20770383
- Volume :
- 11
- Issue :
- 23
- Database :
- Academic Search Index
- Journal :
- Journal of Clinical Medicine
- Publication Type :
- Academic Journal
- Accession number :
- 160738540
- Full Text :
- https://doi.org/10.3390/jcm11237068