Study objectives: Diagnosis of acute left ventricular failure (LVF) is often difficult in the prehospital setting. Brain natriuretic peptide (BNP) is a marker of LVF. The object of this study was to evaluate the feasibility of BNP measurement during the prehospital management of patients with dyspnoea.
Design: Prospective feasibility study, in the Paris Emergency Medical Service (SAMU).
Patients: All patients, aged 50 years and over, presenting with acute dyspnoea were included in the study, unless the dyspnoea was of circumstantial origin. Bedside BNP assays were conducted in parallel with the usual clinical management. For each patient, three diagnoses (cardiac, respiratory or uncertain) were established: firstly, according to the usual clinical criteria (diagnosis 1); secondly (diagnosis 2) according to the result of BNP measurement. When the diagnoses 1 and 2 were not in agreement, patients were entered into a category labeled "diagnostic correction".
Results: Fifty-two patients were included in the study. Twenty-one patients had clinically obvious LVF (diagnosis 1' = cardiac). For seven other patients, the clinical variables suggested a respiratory cause (diagnosis 1 = respiratory). For 24 patients dyspnoea was due to a non-identified cause (diagnosis 1 = uncertain). BNP levels were measured in 51 out of 52 patients (one failure). Only nine patients had a BNP level lower than the threshold value of 100 pg ml(-1). In 71% the diagnosis 1 was corrected after BNP estimation. Only two of 27 patients with marked bronchospasm had a BNP level lower than 100 pg ml(-1).
Conclusion: Estimation of BNP is both feasible and easy in prehospital care, and can confirm the cardiac origins of atypical acute dyspnoea. In elderly patients LVF appears to be clinically underestimated. BNP assay may produce improvements in prehospital management of patients with dyspnoea.