Document details

Brain natriuretic peptide based strategy to detect left ventricular dysfunction in Chagas disease : a comparison with the conventional approach.

Author(s): Ribeiro, Antônio Luiz Pinho ; Teixeira, Mauro Martins ; Reis, Adelina Martha dos ; Silva, André Talvani Pedrosa da ; Perez, Amanda Arantes ; Barros, Márcio Vinicius Lins ; Rocha, Manoel Otávio da Costa

Date: 2015

Origin: Oasisbr

Subject(s): Natriuretic peptides; Cardiomyopathy; Diagnosis; Ventricles; Chagas disease


Description

Background: Left ventricular dysfunction (LVd) is the main predictor of mortality in Chagas disease (ChD). Aims: To compare the diagnostic performance of the conventional approach (ECG and chest X-ray) in the recognition of LVd in ChD, with a new strategy, in which BNP is measured in patients with an abnormal ECG. Methods: Consecutive ChD patients recruited at an Outpatient Reference Center in Belo Horizonte, Brazil, without other systemic diseases, in 1998–99 (sample 1, n =165) and in 2001–02 (sample 2, n =62) underwent ECG, chest X-ray, BNP measurement and echocardiography. Results: The prevalence of LVd (ejection fraction _0.40) was 9.1% in the sample 1. The conventional strategy recognized all patients with LVd (sensitivity: 100%, 95% CI: 79.6–100% and negative predictive value _PV 100%, 92.1–100%), but with low specificity (30%, 95% CI: 23.2–37.8) and +PV (12.5%, 95% IC: I7.7–19.6). The BNP/ECG strategy showed significantly better specificity (96.0%, 95% CI: 91.5–98.2, p <0.001) and +PV (66.7%, 95% CI: 43.7–83.7, p <0.001), and non-significantly lower sensitivity (80.0%, 95% CI: 54.8–93.0, p =0.25) and _PV (98.0%,95% CI: 94.2–99.3, p =0.08). Overall accuracy was improved with the new strategy. (94.5%,95% CI: 90.0– 97.1_36.4%, 95% CI: 29.4–43.9, p <0.001). Similar results were obtained for the sample 2. Conclusions: The BNP-based strategy was more accurate than the conventional approach in the detection of LVd in ChD patients and should be considered as a valid option.

Document Type Journal article
Language English
facebook logo  linkedin logo  twitter logo 
mendeley logo

Related documents

No related documents