Peak BNP >21,650 pg/mL Predicts ECMO Need in Pediatric Fulminant Myocarditis

Fulminant myocarditis (FM) in children is a severe, life-threatening condition characterized by acute inflammation of the heart muscle, leading to rapid cardiac dysfunction and cardiogenic shock. Early and aggressive intervention, often including extracorporeal membrane oxygenation (ECMO), is crucial for survival. However, identifying children who will benefit most from ECMO in a timely manner remains a significant clinical challenge. Current diagnostic and prognostic indicators can be inconsistent, highlighting an urgent need for reliable early warning markers to guide critical care decisions and improve patient outcomes.

Researchers retrospectively analyzed clinical data from 22 children diagnosed with fulminant myocarditis (FM) admitted to a Pediatric Intensive Care Unit between June 2019 and June 2025. Patients were categorized into an ECMO group (n=10) and a non-ECMO group (n=12), as well as a survival group (n=17) and a death group (n=5). Data collected included demographics, laboratory values (e.g., BNP, lactate), echocardiographic measurements (e.g., LVEF, FS), and treatment details (e.g., VIS). The predictive value of early warning indicators for ECMO initiation was assessed using receiver operating characteristic (ROC) curve analysis.

Children in the ECMO group exhibited significantly lower left ventricular ejection fraction (LVEF) and fractional shortening (FS), alongside higher peak brain natriuretic peptide (BNP), peak vasoactive-inotropic score (VIS), and peak lactate levels compared to the non-ECMO group (all P<0.05). They also showed a higher proportion of interventricular septal and ventricular wall hypokinesis. ROC analysis identified a critical threshold:

Peak BNP >21,650 pg/mL predicted the need for ECMO with a sensitivity of 80%, a specificity of 82%, a positive predictive value of 80%, a negative predictive value of 60%, and an area under the curve of 0.771 (P<0.05). Overall mortality in the cohort was 23% (5/22). Non-survivors had significantly lower Pediatric Critical Illness Score (PCIS), LVEF, and FS, and higher peak VIS, peak lactate, and lactate levels at 24 hours post-treatment, as well as a higher rate of ECMO use (all P<0.05).