A prospective study of biopsy-proven myocarditis: prognostic relevance of clinical and aetiopathogenetic features at diagnosis.

AIMS Myocarditis may be idiopathic, viral, and/or immune; frequency of these forms and prognosis are ill-defined. We aimed at identifying aetiopathogenetic and prognostic markers in myocarditis, including viral genome on endomyocardial biopsy (EMB) by polymerase chain reaction (PCR) and serum anti-heart autoantibodies (AHA). METHODS AND RESULTS We studied 174 patients, 110 males, aged 36 +/- 18 years, median follow-up 23.5 months, range 10-54; 85 patients had active myocarditis and 89 borderline myocarditis (no diffuse or severe inflammation) (Dallas criteria). Serum AHA were detected by indirect immunofluorescence. PCR was used to detect virus. Six-year actuarial survival was 73%. AHA were found in 56% of patients and positive PCR in 26%. Univariate predictors of death/transplantation were young age, longer symptom duration, giant cell myocarditis, NYHA II-IV, positive PCR, presentation with LV dysfunction, clinical signs/symptoms of heart failure, and echocardiographic and haemodynamic indexes of cardiac dysfunction. By Cox univariate analysis, highest risk was conferred by clinical signs/symptoms of left (HR = 4.3, CI 1.7-10.8, P = 0.002) and right heart failure (HR 3.4, CI 1.5-7.3, P = 0.002). CONCLUSION In myocarditis, biventricular dysfunction at diagnosis was the main predictor of death/transplantation. AHA identified immune-mediated myocarditis in the majority of cases. Viral genome was a univariate predictor of adverse prognosis. Our approach of using AHA and positive PCR as aetiopathogenetic markers should help patient selection and recruitment in future studies on aetiological therapy.