Measurement, consequences and determinants of time to diagnosis in children with new-onset heart failure: A population-based retrospective study (DIACARD study)

Background: Time from first symptoms to diagnosis, called time to diagnosis, is related to prognosis in several diseases. The aim of this study was to assess time to diagnosis in children with new-onset heart failure (HF) and assess its consequences and determinants.

Methods: A retrospective population-based observational study was conducted between 2007 and 2016 in a French tertiary care center. We included all children under 16 years old with no known heart disease, and HF confirmed by echocardiography. With logistic regression used for outcomes and a Cox proportional-hazards model for determinants, analyses were stratified by HF etiology: congenital heart diseases (CHD) and cardiomyopathies/myocarditis (CM).

Results: A total of 117 children were included (median age [interquartile range (IQR)] 25 days (6-146), 50.4% were male, 60 had CHD and 57 had CM). Overall median (IQR) time to diagnosis was 3.3 days (1.0-21.2). The frequency of 1-year mortality was 17% and 1-year neuromotor sequel 18%. Death at 1 year was associated with low birth weight for all patients (adjusted odds ratio 0.24, 95% confidence interval [CI] 0.08-0.68) and time to diagnosis below the median with CM (0.09, 0.01-0.87) but not time to diagnosis above the median for all patients (0.59, 0.13-2.66). Short time to diagnosis was associated with clinical severity on the first day of symptoms for all patients (adjusted hazard ratio 3.39, 95% CI 2.01-5.72), and young age with CM (0.09, 0.02-0.41).

Conclusions: In children with new-onset HF presenting in our region, median time to diagnosis was short. Long time to diagnosis was not associated with poor outcome.