Sonographic chest B-lines anticipate elevated B-type natriuretic peptide level, irrespective of ejection fraction

Acute pulmonary edema is a common problem facing emergency department (ED) physicians, and a percentage of these patients are admitted to a CCU. The diagnosis of acute pulmonary edema remains a challenge for the following reasons: the presentation could be in combination with other diseases, such as chronic obstructive airway disease; and other diseases have a presentation that is similar to that of acute pulmonary edema.

Cardiologists and intensivists commonly assess the heart using echocardiography. To save time, an extended evaluation could be performed using the same probe to complete the evaluation without changing the probe. Chest ultrasound is used to detect subpleural interstitial edema lines (B-lines) and pleural effusion; this method is used to exclude other important diseases, e.g., pneumothorax.

The assay for plasma ProBNP is a useful test for the evaluation of patients with dyspnea, and it is particularly useful as a component of the evaluation of a suspected heart failure when the diagnosis is uncertain [1].

The correlation between ProBNP and B-lines was studied using a linear probe. To simplify and expedite the procedure, we correlated the B-lines detected by the echocardiography probe to the NT-proBNP assay.

We enrolled 61 patients over a period of 6 months. The patients provided signed consent. The mean age was 66.8 years, with a range of 44–93 years; 49.2 % of the subjects were male. A total of forty-seven of the 61 patients had B-profile and hemodynamic pulmonary edema. The clinical characteristics of the patients in relation to ultrasound chest profiles are shown in Table 1.

The median NT-proBNP levels in patients with B-profile was 6200, compared with a median of 180 in the subjects with A-profile (CI 0.33–0.82).

The distributions in the 2 groups differed significantly (Mann–Whitney U-static 3.98). Based on the threshold level of NT-proBNP of 250, the sensitivity of detecting B-profile on ultrasound was 95.0 %, and the specificity was 92.0 %. The positive predictive value of the B-profile was 90.0 %, and the negative predictive value was 88.0 % (Table 2).

The systolic function in the subjects with a B-profile was below 50 % in 84.3 % of cases and normal in 15.7 % of cases. The subjects with an A-profile all had systolic function >55 %.

Of those subjects with a B-profile, 94 % had a Framingham score of CHF >4; all of the subjects with an A-profile had a score of <4, p < 0.0001. National Health and Nutrition Examination Survey (NHANES) scores of >3 were found in 96 % of those with a B-profile, and all of the subjects with an A-profile had scores <3, p < 0.0001.

Lung ultrasound is becoming an essential tool for the diagnosis of pulmonary disease irrespective of the echocardiographic findings. This method is becoming a standard method to complement conventional Doppler echocardiography in the rapid evaluation of patients presenting with dyspnea in the emergency department (for the differential diagnosis of dyspnea), in hospital management (for serial evaluations in the same patient and for tailoring diuretic therapy), in the prehospital emergency setting (with hand-held echocardiography), and in the stress echocardiography lab (as a sign of acute pulmonary congestion during stress) [4].

Providing a reliable and repeatable estimation of EVLW, B-profile assessment by lung ultrasound represents a new, helpful tool for the cardiologist; this tool could be employed at all stages of the management of heart failure patients and could be used in the differential diagnosis of dyspnea [4, 6].

Increased left ventricular filling pressure is a common hemodynamic trigger for natriuretic peptides and B-profile. Wall distension is generally considered the main mechanical stimulus for natriuretic peptide production by ventricular tissue from stretched cardiomyocytes [7]. The use of NT-proBNP and B-type natriuretic peptide at the rule-out threshold recommended by the recent European Society of Cardiology guidelines on heart failure provides excellent ability to exclude acute heart failure in the acute setting with reassuringly high sensitivity [8]. The specificity of the natriuretic peptides is modest and variable, and therefore confirmatory diagnostic testing by cardiac imaging is required in the case of positive results [8].

The presence of B-profile as a sign of pulmonary interstitial edema is linked to augmented left ventricular filling pressures, which unbalance Starling forces at the alveolar–capillary barrier, resulting in pulmonary congestion [9]. Therefore, the overall good concordance and significant correlation between B-lines and NT-proBNP found in this study are not surprising.

The presence of B-profile correlating with higher BNP levels were previously studied using a linear probe because this method is easily learned and reproducible [10]. Although earlier literature identified similar results with curved probes [11], we used a curved probe in this study to allow for extended echocardiography examinations of the patient’s heart and chest using the same probe. The use of an echocardiography probe in detecting B-profile in the bedside evaluation of patients with known or suspected heart failure by experienced and inexperienced echocardiologists was previously studied and shown to be a reliable diagnostic tool [12]. Cardiac probes can show lung artifacts, may be with a quality inferior to some other equipment. We used an 8-zone scanning protocol compared with an NT-proBNP, and we detected that minimal pleural effusion corresponded more closely with heart failure in cases in which the patient presented with acute dyspnea. An eight-lung window protocol was used in the studies by Liteplo et al [11] and Volpicelli et al [2]. Lichtenstein et al. used a six-window protocol in previous studies and included the PosteroLateral Alveolar and/or Pleural Syndrome (PLAPS)-point, a posterior area accessible in supine patients, locating all free effusions, regardless of their volume [4].

In the literature, there have not been stringent criteria defining how many B-lines constitute a significant lung ultrasound finding. Lichtenstein et al. first used the presence of at least three B-lines per field of scan seen longitudinally between two ribs, with a distance between two B-lines <7 mm, as criteria for abnormality, and the majority of authors have followed this convention (1, 2, 10, 13). These criteria have been used irrespective of the scan technique, whether the scan is transverse in the intercostal space or longitudinal across ribs. Typically, a micro convex or phased array probe with a narrow footprint is used. Some operators decide to use a linear or curvilinear probe with a broad footprint. These probes allow for views across several rib spaces or for a longer view of the pleural line. A positive scan with these probes should be similar to the general convention of at least three B-lines less than 7 mm apart [14].

The presence of B-profile is specific for an elevated NT-proBNP level although the sensitivity is variable [15]. In a linear probe study, the sensitivity was 33.3 %. B-lines had a sensitivity of 100 % and a specificity of 92 % in the diagnosis of pulmonary edema when compared with COPD [16]. Gargani L et al. used a cardiac probe and determined that B-lines are reliable in predicting the cardiogenic origin of dyspnea, with an accuracy comparable to natriuretic peptides [13].

The echocardiographic data appear useful in the context of a lung ultrasound examination, given the high accuracy published in previous articles (Sensitivity 97 %, specificity 95 %) [3]. B-profile could be a plausible alternative in acute settings where natriuretic peptide analysis is not available or when there is insufficient time to perform the assay, as occurs in patients with rapidly developing acute respiratory failure. The estimated time in getting the result of NT-proBNP after sampling of the patient is almost 3 h in our institute.

The presence of B-profile correlated well with the clinical scores of heart failure in our study. A total of 94 % of the patients with a B-profile had a Framingham score of CHF >4, and 96 % had NHANES scores of >3. All of the patients with A-profile had normal scores. Our study demonstrates the clinical correlation of chest ultrasound examinations with heart failure scores, as showed previously in the BLUE-protocol study [3].

Chest sonography focused on B-profile when combined with a comprehensive echocardiography study could evaluate systolic and diastolic cardiac function [15]. Patients with cardiogenic edema are more likely to have a lower ejection fraction and a higher degree of diastolic dysfunction [17]. In our study, 15.4 % of patients with cardiogenic pulmonary edema and B-profile had a left ventricular ejection fraction (LVEF) >50 % with high NT-proBNP levels, predominantly because of diastolic dysfunction.

Limitations of the study include low number of patients recruited. This was due to two main reasons; NT-proBNP was ordered by the ED physician on difficult patients with high heart failure scores, thus missing patients with genuine hemodynamic pulmonary edema. Thus, other profiles of BLUE-protocol in our population were not included. Furthermore, patients incapable of providing informed consent, even though having pulmonary edema, were omitted from the study. We compared the B-profile to the clinical signs of heart failure and to a correlation with natriuretic peptides. Spectral tissue Doppler-derived index(E/E’) for assessing left atrial pressure as well as patients having heart failure with preserved ejection fraction were not studied, thus considered a limitation to our study.

Detecting the B-profile is highly sensitive and specific for elevated NT-proBNP helping in diagnosing pulmonary edema. A cardiac probe (curvilinear 5 MHz) can be used, although of resolution inferior to some micro convex probes. Performing chest ultrasound could be part of the echocardiography evaluation in patients with acute dyspnea.