Diagnostic yield of etiologic workup in acute myocarditis: insights from the AMPHIBIA Registry
A Villa, M Proust, G Hekimian, F Cohen, E Gandjbakhch, J Proukhnitzky, A Redheuil, F Huang, M Zeitouni, M Papo, E Vicaut, J Silvain, G Montalescot, A Combes, M KerneisAbstract
Background
Acute myocarditis is a heterogeneous condition with a wide range of potential infectious, and/or immune-mediated causes. While recent guidelines have questioned the routine use of systematic etiological investigations in all patients, comprehensive non-invasive work-ups remain frequently performed. However, the diagnostic yield of such systematic strategies and their impact on therapeutic decision-making and prognosis remains insufficiently defined. This study aimed to assess the diagnostic performance and the prognostic implications of a standardized, non-invasive etiological assessment in acute myocarditis patients.
Methods
We analyzed consecutive patients enrolled in the AMPHIBIA registry between 2019 and 2024 with confirmed acute myocarditis, diagnosed either by endomyocardial biopsy (EMB) or cardiac magnetic resonance (CMR). Both fulminant and non-fulminant forms were included. All patients underwent a systematic and extensive etiological evaluation, comprising detailed microbiological screening and comprehensive autoimmune blood testing. For each testing panel, we quantified the rate of testing, analytical positivity, and final etiological yield. The primary composite endpoint included all-cause mortality, need for temporary mechanical circulatory support, heart transplantation, or occurrence of severe ventricular arrhythmia. Logistic regression models were used to determine whether clinical infectious prodromes and routinely available biomarkers could predict microbiological findings considered etiologically contributory.
Results
A total of 236 patients were included during the study period. The median age was 32 years, and 69% were male and 34% were hospitalized in ICU. Overall, 59% of patients had at least one positive microbiological result; however, only 27% were considered etiologically contributory, and merely 18% led to a modification in treatment. Viral serologies showed limited clinical utility, whereas nasopharyngeal PCR testing appeared more informative. A similar discrepancy was observed for autoimmune testing: among 203 tested patients, 56% had at least one positive result, but only 6% were judged etiologically relevant and treatment-modifying. The primary endpoint occurred among 21% of our population, with a median follow-up of 1.4 years. Patients with a completely negative etiological work-up demonstrated the most favorable outcomes. In multivariable analysis, the combination of simple infectious prodromes and routine biomarkers predicted contributory microbiological findings with good discrimination (AUC 0.84).
Conclusions
In this contemporary cohort of acute myocarditis, systematic infectious and autoimmune blood testing was common but provided limited additional etiological and therapeutic benefit. Importantly, straightforward clinical features may help refine patient selection and better target microbiological investigations toward those most likely to derive meaningful diagnostic benefits.