Can plasma levels of N-terminal pro-brain natriuretic peptide be useful in predicting the risk of developing chronic pulmonary hypertension in patients with pulmonary embolism?

Abstract

We read with great interest the manuscript by Dentali and colleagues [1] on the role of N-terminal pro-brain natriuretic peptide (Nt-proBNP) as a biomarker of pulmonary hypertension in patients with pulmonary embolism (PE). Risk stratification of patients with PE has recently received considerable attention because the availability of specific cardiac biomarkers might facilitate the identification of patients at high risk of shortor long-term complications and improve therapeutic decision making. Pro-hormone BNP is a 108 amino acid peptide that is synthesized in the myocytes, and is cleaved by an endoprotease into two polypeptides: the inactive Nt-proBNP and the biologically active BNP. The secretion and release of the pro-hormone BNP is enhanced during the hemodynamic stress that is associated with ventricle dysfunction due to volume expansion and pressure overload. Chronic thromboembolic pulmonary hypertension (CTPH) is a relatively common, serious complication of PE, whose early identification is problematic [2]. Dentali et al. [1], in their study of 49 patients with PE, find that there is a correlation between the plasma levels of Nt-proBNP and pulmonary artery systolic pressures (PAPs), and that low plasma Nt-proBNP levels have high negative predictive value (97.2%) for CTPH, suggesting that measurement of plasma Nt-proBNP may be useful to rule out preclinical or symptomatic CTPH. This manuscript is original and interesting; however, some methodological flaws may limit its clinical relevance. As acknowledged by the authors [1], one of the limitations is that the study population was small and heterogenous in terms of baseline characteristics, time from the index event and type and duration of antithrombotic treatment. As a matter of fact, the plasma levels of Nt-proBNP are influenced by several physiological conditions and disease states [3], including age, gender, ethnicity, obesity, renal impairment and the severity of PE. Unfortunately, the sample size of this study [1] was not large enough to weigh each of these confounding factors in a multivariate analysis. In addition, since patients develop CTPH approximately 2 years after a first episode of symptomatic PE [2], in order to ascertain a possible role of BNP on early diagnosis of CTPH, the follow-up of the studied patients should have been longer and less variable. Finally, the test is limited by a high intra-individual variability in the plasma levels of Nt-proBNP, associated with the circadian rhythm of hemodynamic indices, physical activity and fluid intake [3], which may increase the chance of false-positive or false-negative results in each individual patient. Moreover, studies of different clinical populations identify different cut-off concentrations of Nt-proBNP [4, 5]; therefore, the optimal cut-off of plasma Nt-proBNP levels for assessing their negative predictive value for CTPH needs to be established, while the individual Nt-proBNP concentrations should be considered cautiously. In conclusion, the results of the study by Dentali et al. [1] are certainly useful in generating a hypothesis of a possible role for Nt-proBNP plasma levels in predicting long-term complications of PE, but they should be considered preliminary. This intriguing hypothesis deserves to be tested in well-designed and prospective studies to clarify whether the Nt-proBNP plasma levels could indeed be useful in clinical practice. F. Lussana (&) G. M. Podda M. Cattaneo Divisione di Medicina Generale III, Dipartimento di Medicina, Chirurgia e Odontoiatria, Azienda Ospedaliera San Paolo, Universita di Milano, Via A. di Rudini, 8, 20142 Milan, Italy e-mail: federico.lussana@ao-sanpaolo.it