- Email: [email protected]
Diagnostic role of plasma BNP levels in neonates with signs of congenital heart disease
International Journal of Cardiology 147 (2011) 42–46
Contents lists available at ScienceDirect
International Journal of Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j c a r d
Diagnostic role of plasma BNP levels in neonates with signs of congenital heart disease Periklis A. Davlouros a,⁎, Ageliki A. Karatza b, Ioanna Xanthopoulou a, Gabriel Dimitriou b, Aikaterini Georgiopoulou a, Stefanos Mantagos b, Dimitrios Alexopoulos a a b
Department of Cardiology, Patras University Hospital, Greece Department of Paediatrics, Patras University Hospital, Greece
a r t i c l e
i n f o
Article history: Received 29 May 2009 Accepted 24 July 2009 Available online 26 August 2009 Keywords: Plasma BNP Congenital heart disease Neonates
a b s t r a c t Background: Only a few studies have examined the relationship of plasma BNP levels and congenital heart disease (CHD) in neonates and these mainly concern preterm neonates with patent ductus arteriosus. We aimed to investigate the diagnostic role of plasma BNP in neonates admitted in the neonatal intensive care unit, (NICU), with signs of congenital heart disease (CHD). Methods: Prospective assessment of plasma BNP levels in 75 consecutive neonates with suspected CHD (heart murmur, respiratory distress, or cyanosis), admitted in the NICU of our university hospital. The final diagnosis was done with echocardiography. Results: Haemodynamically significant Left to Right shunts, (hsLtR), were found in 29 neonates, insignificant LtR shunts in 22, no heart disease in 15 and cyanotic heart disease in 9. BNP levels were significantly higher in neonates with hsLtR shunts vs. all other groups (logBNP 2.9 ± 0.5 pg/ml vs. 1.5 ± 0.4 pg/ml vs. 1.5 ±0.3 pg/ ml vs. 1.6 ± 0.2 pg/ml, p b 0.0001). Plasma BNP levels N 132.5 pg/ml had 93.1% sensitivity and 100% specificity for diagnosing hsLtR shunts (accuracy 99.6%). Conclusions: Plasma BNP is a reliable test for diagnosing hsLtR shunts in the NICU. This will alert the neonatologist for ordering an echocardiographic examination, or if the latter is not available, for transferring the neonate to an appropriate tertiary centre with neonatal-paediatric cardiology facilities. Normal BNP levels imply the absence of a significant LtR shunt, but may not exclude cyanotic heart disease. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The overall prevalence of cardiovascular malformations at birth is approximately 4–6 per 1000 live births [1]. Nowadays many of them are diagnosed during intrauterine life, however a substantial proportion of neonates with congenital heart disease (CHD) are diagnosed after birth, either in neonatal intensive care units (NICU), or in and out of hospital basis after echocardiographic examination. The most frequent indications for an echocardiogram in the NICU are a heart murmur, respiratory distress and cyanosis [2]. However echocardiography is not always available in the NICU and when available, a properly trained sonographer in the diagnosis of CHD is also needed. Plasma brain natriuretic peptide (BNP), is a well known accurate and easy to conduct bedside laboratory test for the diagnosis and guidance of heart failure therapy in adults [3–5]. Unfortunately only a few studies have examined the relationship of plasma BNP levels and CHD
⁎ Corresponding author. Patras University Hospital, Rion, Patras, 26500, Greece. Tel.: +30 2610999281; fax: +30 6932468624. E-mail address: [email protected] (P.A. Davlouros). 0167-5273/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2009.07.029
in neonates and children [6–11]. It has been shown that plasma BNP levels are a useful adjunct to echocardiography in the diagnosis of patent ductus arteriosus and its response to treatment in premature neonates and in the diagnosis of persistent pulmonary hypertension of the newborn [6,10,12–15]. There is only one recent study showing that the use of a rapid bedside plasma BNP assay may be useful for the diagnosis of cardiovascular problems in neonates with respiratory distress during the first few days of their life in the NICU [10]. We aimed to investigate any possible diagnostic role of plasma BNP levels in neonates admitted in our NICU, with signs suggestive of heart disease. 2. Methods We examined 75 consecutive neonates which were admitted in our NICU, (Patras University Hospital), with signs suggestive of CHD. More specifically, a heart murmur, respiratory distress and/or cyanosis, were considered indices of possible CHD. The existence of the latter was ruled in or out with an echocardiogram. A haemodynamically significant Left to Right (hsLtR) shunt was defined as a Qp/Qs ≥ 1.5 in case of an atrial or ventricular septal defect (ASD, VSD). In case of a patent ductus arteriosus (PDA), a hsLtR shunt was defined as the existence of at least one of three clinical signs of severity (murmur, increased precordial activity, bounding pulses) and the coexistence of at least one of two echocardiographic criteria of severity, namely a left atrial/aortic root diameter (LA/Ao) N 1.5 and/or a LtR shunt with a pulsatile Doppler flow pattern at the level of the
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
43
Table 1 Echocardiographic diagnosis, presenting signs and plasma BNP levels (pg/ml) in neonates with hsLtR shunts. Echocardiographic diagnosis
Presenting sign
Plasma BNP levels (pg/ml)
Echocardiographic diagnosis
Presenting sign
Plasma BNP levels (pg/ml)
PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA
Murmur RD Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur
5500 4920 4550 3798 3180 2360 2110 1830 1670 1220 651 290 277 116
PDA, ASD PDA, ASD PDA, ASD PDA, ASD PDA, ASD PDA, ASD, VSD ASD, VSD ASD, VSD ASD, VSDs ASD, VSD ASD, VSD ASD, VSD VSD VSD VSD
RD Murmur Murmur Murmur Murmur RD Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur
1290 1260 912 673 98.9 344 906 800 771 500 376 248 563 155 139
hsLtR shunt: haemodynamically significant Left to Right shunt. PDA: patent ductus arteriosus, RD: respiratory distress, ASD: atrial septal defect, and VSD: ventricular septal defect.
PDA [16]. Neonates with birth asphyxia, persistent pulmonary hypertension, or non congenital heart disease were excluded. Neonates were classified in 4 groups. Neonates with hsLtR shunts as defined above, neonates with haemodynamically insignificant LtR shunts (Qp/Qsb 1.5) and no significant PDA, neonates without heart disease and neonates with cyanotic heart disease. Plasma BNP levels were determined quantitatively in EDTAanticoagulated whole blood with a rapid bedside immunofluorometric assay kit (Triage BNP; Biosite Diagnostics, San Diego, CA) [15]. Blood samples were collected either before or after the echocardiogram and BNP measurement was conducted immediately. The exact timing of blood sampling is described in the Results section. Kolmogorov–Smirnov test showed that BNP levels, (pg/ml), were skewed and therefore results are expressed as medians (range) or were logarithmically transformed and expressed as mean ± SD. The Mann–Whitney U test and the Kruskal–Wallis test were used to assess differences of BNP levels between two or more than two groups respectively. The Student's t-test for independent samples and one-way ANOVA with Bonferroni post-hoc comparisons were used when BNP levels were logarithmically transformed. ROC curve analysis was used to examine the ability of BNP levels to detect significant PDA in premature neonates. All tests were two-tailed and statistical significance was considered for p-values b 0.05. All statistical analyses were performed using SPSS for windows (version 15.0 SPSS Inc., Chicago, IL, USA).
The study was approved by the ethics committee of our hospital and all parents gave informed consent for blood sampling of their neonates.
3. Results Of the 75 neonates studied, 34 (45.3%) were premature (b37 weeks gestational age, birth weight 1478.9 ± 653.2 g) with 10 of them b28 weeks gestational age (13.3% of total, birth weight 872 ± 112.1 g). Forty-one (54.7%) were normal term neonates (birth weight 3104 ± 410 g). The most frequent reason for conducting an echocardiogram in the NICU was the existence of a heart murmur (75%), followed by respiratory distress (13%) and cyanosis (12%). All but one neonate had a normal left ventricular (LV) fractional shortening and ejection fraction (EF), (35.3 ± 4.5% and 68.5 ± 6.3% respectively). Twenty-nine neonates (38.6%), had a hsLtR shunt, 22 (29.3%) had a haemodynamically insignificant LtR shunt, 15 (20%) did not have
Table 2 Echocardiographic diagnosis, presenting signs and plasma BNP levels (pg/ml) in neonates without hsLtR shunts. Echocardiographic diagnosis
Presenting sign
Haemodynamically insignificant LtR shunt ASD Murmur ASD Murmur ASD Murmur ASD Murmur ASD Murmur ASD Murmur ASD Arrhythmia ASD, dextrocardia Murmur VSD Murmur VSD Murmur VSD Murmur VSD, PFO Murmur VSD, PFO Murmur VSD, ASD Murmur VSD, ASD Murmur VSD, ASD Murmur VSD, ASD Murmur PDA Murmur PDA Murmur PDA Murmur PDA RD PDA, ASD RD
BNP levels (pg/ml) 12.5 5 14.4 27.6 100 45 48 23.5 48.4 71.1 45.8 75 25.4 6.5 15.9 7.5 126 25.6 93.5 46 65 82.2
Echocardiographic diagnosis Cyanotic heart disease DORV, VSD TGA, ASD TGA, VSD TGA, VSD TAPVR, PDA TAPVR Critical PS, ASD PA-VSD PA-VSD No heart disease Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal
Presenting signs
BNP levels (pg/ml)
Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis
95 35.9 26.4 35 45 39.1 29.8 54.5 120
RD RD RD RD RD Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur
56.9 55 74.7 29.3 10.1 11.3 36.2 14 83 8.9 16.6 44.1 42 50.2 14.8
PDA: patent ductus arteriosus, RD: respiratory distress, ASD: atrial septal defect, VSD: ventricular septal defect, DORV: double outlet right ventricle, TGA: transposition of the great arteries, TAPVR: total anomalous venous return, PS: pulmonary stenosis, and PA: pulmonary atresia.
44
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
Fig. 1. Plasma BNP levels in relation to the underlying echocardiographic diagnosis. From 75 consecutive neonates with signs of heart disease, neonates with hsLtR shunts had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (ANOVA p b 0.0001). (hsLtR shunt: haemodynamically significant Left to Right shunt and CHD: congenital heart disease).
heart disease and 9 (12%) had cyanotic heart disease (Tables 1 and 2). Blood sampling was done within 12 h from birth in 9 neonates with cyanosis and in one neonate with heart failure signs (27 weeks gestational age with a massive PDA, EF 40%). Blood sampling in the remaining neonates was done beyond the second day of life. 3.1. BNP levels in neonates with hsLtR shunts Plasma BNP levels were significantly higher in neonates with hsLtR shunts compared to the other three groups (median BNP 800 pg/ml vs. 45.4 pg/ml vs. 36.2 pg/ml vs. 39.1 pg/ml p b 0.001, logBNP 2.9 ± 0.5 pg/ ml vs. 1.5 ± 0.4 pg/ml vs. 1.5 ± 0.3 pg/ml vs. 1.6 ± 0.2 pg/ml ANOVA
Fig. 3. Plasma BNP levels in relation to the underlying echocardiographic diagnosis. From 75 consecutive neonates with signs of heart disease, neonates with hsPDA had significantly higher plasma BNP levels compared to neonates with hsLtR shunts other than PDA (p = 0.016). Both groups had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (ANOVA p b 0.0001). (hsPDA: haemodynamically significant patent ductus arteriosus, hsLtR shunt: haemodynamically significant Left to Right shunt, and CHD: congenital heart disease).
p b 0.0001, Bonferroni p b 0.0001), (Fig. 1). ROC curve analysis showed that plasma BNP levelsN 132.5 pg/ml had 93.1% sensitivity and 100% specificity for diagnosing hsLtR shunts and BNP levels b 96.9 pg/ml had 100% sensitivity and 93.5% specificity for ruling out hsLtR shunts. The overall test accuracy was 99.6% (Fig. 2). When excluding extremely premature neonates (b28 weeks, N = 10) from analysis, the results did not change. Plasma BNP levels did not correlate significantly with the EF in the total population of the study. When examining only neonates with hsLtR shunts, there was a marginal negative correlation between BNP and EF (r = −0.47, p = 0.05). 3.2. BNP levels in neonates with hsPDA Neonates with hsPDA (N = 19) had significantly higher plasma BNP levels compared to the rest of neonates with hsLtR shunts (N = 10), (logBNP 3.06 ± 0.5 pg/ml vs. 2.6 ± 0.3 pg/ml p = 0.016). Neonates with hsPDA also had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (logBNP 3.06 ± 0.5 pg/ml vs. logBNP 1.5 ± 0.4 pg/ml vs. 1.5 ± 0.3 pg/ml vs. 1.6 ± 0.2 pg/ml ANOVA p b 0.001, Bonferroni p b 0.0001), (Fig. 3). ROC analysis showed that a BNP N 201.5 pg/ml had 89.5% sensitivity and 100% specificity for diagnosing a hsPDA and a BNPb 96.9 pg/ml had 100% sensitivity and 93.5% specificity for ruling out a hsPDA. 3.3. BNP levels in neonates with hsLtR shunts other than PDA
Fig. 2. ROC analysis. Plasma BNP levels N 132.5 pg/ml had 93.1% sensitivity and 100% specificity for diagnosing hsLtR shunts. The overall test accuracy, (area under the curve), was 99.6%. (hsLtR shunt: haemodynamically significant Left to Right shunt).
Neonates with hsLtR shunts other than PDA, (i.e. ASD and VSD, N = 10), had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (logBNP 2.6 ± 0.3 pg/ml vs. logBNP 1.5 ± 0.4 pg/ml vs. 1.5 ± 0.3 pg/ml vs. 1.6 ± 0.2 pg/ml ANOVA p b 0.001, Bonferroni p b 0.0001), (Fig. 3). A BNP of 147 pg/ml had 100% sensitivity and specificity in this group.
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
3.4. BNP levels in premature neonates with hsLtR shunts When examining only premature neonates (N = 34), 20 of them (59%) had a hsLtR shunt, most frequently PDA (N = 16). The latter had significantly higher BNP levels compared to neonates with insignificant shunts and neonates without heart disease (logBNP 2.9 ± 0.56 pg/ml vs. logBNP 1.4 ± 0.4 pg/ml vs. 1.3 ± 0.3 pg/ml ANOVA p b 0.001, Bonferroni p b 0.0001). In premature neonates with a significant PDA a BNP N 283.5 pg/ml had 81.3% sensitivity and 100% specificity for diagnosing a significant PDA and a BNP of 193.5 pg/ml had 85% sensitivity and 100% specificity for diagnosing any hsLtR shunt. Ten premature neonates had a gestational age b 28 weeks. From these, 2 had an insignificant PDA (BNP 65 and 93.5 pg/ml) and 8 had hsPDA on echocardiographic examination. In the latter, BNP ranged from 116 to 5500 pg/ml (median 2814 pg/ml) and all of them needed intervention for duct closure, mainly indomethacin therapy. 4. Discussion B-type natriuretic peptide (BNP) is a hormone secreted by the ventricles under haemodynamic stress associated with ventricular enlargement with or without ventricular systolic dysfunction [17]. This study aimed to investigate the diagnostic accuracy of plasma BNP levels in neonates with suspected CHD. Echocardiography is the main diagnostic tool in this setting, however the echocardiographic equipment or a properly trained sonographer in the diagnosis of CHD may not be available in the NICU. In the present study we showed that bedside rapid BNP assessment may aid in the diagnosis or exclusion of significant LtR shunts with volume loading. However, it may not be helpful in cases of cyanotic CHD. 4.1. BNP levels in neonates with hsLtR shunts Plasma BNP levels are elevated in healthy newborns right after birth and then rapidly decrease to b100 pg/ml on day two and b50 pg/ml beyond the fourth day of life [18]. This may reflect volume loading as a result of the transition from fetal to neonatal circulation, or diminished clearance of the peptide from the placenta. We measured plasma BNP levels after the second day of life, except in neonates with cyanosis in whom a rapid and prompt diagnosis is mandatory. We found that overall neonates with a volume loading LtR shunt have significantly elevated plasma BNP levels compared to neonates with haemodynamically insignificant LtR shunts. The latter had BNP levels comparable to neonates without evidence of CHD on echocardiography. The observation that plasma BNP levels are elevated in neonates with hsLtR shunts compared to the rest in our study probably reflects the mechanism of BNP secretion by the ventricles when they are under haemodynamic stress i.e. released by the ventricular myocytes in response to volume and possibly pressure load. Our results are in accord with the few studies which have examined the relationship of plasma BNP and LtR shunting, mainly in children. Plasma BNP levels have been found to be elevated in children with LV and RV volume overload (ASD, VSD, and PDA) without ventricular dysfunction and to correlate significantly with Qp/Qs and LV and RV end-diastolic volumes [8,19]. It has also been shown that in children with LtR shunts plasma BNP levels correlate with shunt volume and systolic RV pressure [9]. Elevated plasma BNP levels have been found in children with CHD and reduced LV systolic function [20]. A negative correlation between plasma BNP levels and LV systolic function has also been described in paediatric patients with CHD in a study including patients with impaired LV systolic function [9]. We also found a weak negative correlation between LV ejection fraction and plasma BNP levels in neonates with hsLtR shunts, extending the above findings to the neonatal population with CHD. However, as expected, the LV function was normal in our population, as LtR shunts in paediatric patients initially lead to right or left ventricular volume overload prior to the emergence of ventricular dysfunction [8]. We conclude that in
45
neonates with hsLtR shunts, plasma BNP levels may reflect the degree of ventricular distension-stretching and secondarily ventricular systolic impairment, (if any), as a result of volume load. This implies that plasma BNP levels may be a useful index of the severity of volume loading in neonates with LtR shunts and may aid in determining the need for an intervention. However, the latter was not examined in our study. There are not many studies examining the diagnostic role of plasma BNP in neonates with suspected CHD. One recent study showed that a bedside plasma BNP assay may be useful for the diagnosis of cardiovascular problems in neonates with respiratory difficulties during the first few days of their life [10]. In this study, plasma BNP levels were significantly higher in neonates with cardiovascular disease compared to the rest with cutoff points of 346, 421, 570.5 and 191.5 pg/pl within 18 h, at 18–36 h, at 36–60 h and beyond 60 h of life. In our study blood sampling was done beyond 48 h for all neonates apart from those with cyanosis and the overall cutoff value was 132.5 pg/ml, 201.5 pg/ml for a hsPDA and 147 pg/ml for hsLtR shunts other than PDA. Our study differs however from that of Ko et al. in that all neonates with suspected heart disease were included and not only those with respiratory problems. 4.2. BNP levels in term and preterm neonates with hsPDA or other hsLtR shunts In contrast to the whole population of neonates admitted in NICUs, there is enough evidence that plasma BNP levels are a useful adjunct to echocardiography in the diagnosis of hsPDA in preterm infants [13,15]. In neonates b28 weeks gestational age a BNP of N300 pg/ml beyond the second day of life, predicts a hsPDA [14]. It has also been shown that a BNP N 550 pg/ml on the second day of life in such neonates predicts interventions for ductus closure (sensitivity 83%, specificity 86%). Successful closure is reflected by a corresponding decrease in BNP and at a cutoff of 70 pg/ml, BNP is a useful screening tool for diagnosis and for monitoring efficacy of treatment of hsPDA [13]. Our study included 10 neonates with a gestational age b 28 weeks and all of them had a PDA on echocardiography. In 8 of them the PDA was haemodynamically significant, all of them needed intervention for ductus closure, mainly indomethacin treatment, and BNP ranged from 116 to 5500 pg/ml, (median 2814 pg/ml), in accord with the above mentioned studies. In the total population examined, neonates with hsPDA had significantly higher plasma BNP levels compared to neonates with hsLtR shunts, neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease. Neonates with hsLtR shunts other than PDA, (i.e. ASD and VSD), had significantly higher plasma BNP compared to the latter three groups (Fig. 3). We might conclude that elevated plasma BNP levels in term or preterm neonates admitted in the NICU with signs suggestive of CHD, imply the existence of a significant volume loading LtR shunt. Preterm infants with a hsPDA have the highest BNP levels, especially those b28 weeks gestational age. The cutoff values of plasma BNP levels for the detection of a hsLtR shunt other than a PDA may be lower. 4.3. BNP levels in neonates with cyanotic CHD Nine neonates (12%), had cyanotic CHD. These neonates did not have elevated plasma BNP levels compared to neonates with insignificant LtR shunts or without heart disease, even though in this group BNP was determined within 48 h from birth, at a time when levels are expected to be relatively elevated [18]. It has been shown that in neonates with transposition of the great arteries (TGA), plasma BNP levels before arterial switch operation are not elevated [21]. On the contrary BNP levels increase beyond 6 h post switch operation in neonates with haemodynamic compromise [21]. In infants and children with significant pulmonary valve stenosis or unoperated tetralogy of Fallot, plasma BNP levels are not elevated compared to healthy controls [9]. Cyanosis itself does not seem to affect plasma BNP levels, as there is no correlation between the latter and oxygen saturation [11]. In another
46
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
study of a very heterogeneous group of children with different types of CHD, there was no correlation between plasma BNP levels and RV outflow obstruction [11]. Data on patients with severe obstruction of the right or left heart demonstrate that not all kinds of increasing ventricular pressure cause directly an increase of plasma BNP, implying that the mechanism of BNP release may be much more complex in CHD [9]. Our study demonstrated a relationship between BNP levels and hsLtR shunts; however a simple correlation between plasma BNP levels and pressure or volume load regardless the underlying specific heart defect may not exist. When considering more heterogeneous populations of CHD patients elevated plasma BNP levels may reflect more accurately the degree of ventricular impairment [9]. Therefore in neonates with cyanotic heart disease without ventricular dysfunction, the haemodynamic stress on the left or right ventricle may not be severe enough for stimulating BNP elevation. We speculate that in these neonates, low plasma BNP levels reflect a compensated status of the heart and cannot preclude any pathology. On the contrary, in adults with repaired cyanotic CHD and heart failure (e.g. tetralogy of Fallot), plasma BNP levels are elevated [22]. Right ventricular dilatation and dysfunction due to pulmonary regurgitation has been shown in these patients and this kind of decompensation may explain the increase in BNP levels [23,24]. Further research is needed on the relationship of cyanotic CHD and BNP levels in neonates. 5. Conclusion Bedside rapid plasma BNP assay is a reliable, relatively accurate and easy to conduct blood test, for diagnosing hsLtR shunts in neonates admitted to the NICU with symptoms and/or signs of suspected CHD. This will alert the neonatologist for ordering an echocardiographic examination, or if the latter is not available, for transferring the neonate to an appropriate tertiary centre with neonatal-paediatric cardiology facilities. When plasma BNP levels are not elevated, there is a certain level of assurance that a haemodynamically significant LtR shunt does not exist, however major cyanotic CHD cannot be excluded. 6. Limitations and strengths of the study Although the number of neonates with significant CHD is adequate in this study, most hsLtR shunts represented PDA. More work is needed to definitively conclude about the role of plasma BNP levels in neonates with other hsLtR shunts like ASD and VSD, as well as its predictive value regarding the need for intervention. However, our study aimed to determine the adjunctive role of BNP in diagnosing CHD in the NICU and in this context we demonstrate some useful information. It is important that this study included only neonates and not a mixed population of infants, children or adults with CHD like all the other studies in the literature. Although our results were in accordance with the above studies to a certain degree, therefore extending their conclusions to the neonatal population, we should stress that data regarding older patients cannot be transferred directly to neonates and BNP cutoff values may be different between neonates and older patients. We consider the results regarding cyanotic CHD very interesting. Certainly the numbers are small, but hypothesis generating and stimulating further research in the field. Finally, haemodynamic data would be desirable in such studies, however this was not possible in our centre.
Acknowledgment The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [25]. References [1] Loffredo CA. Epidemiology of cardiovascular malformations: prevalence and risk factors. Am J Med Genet 2000;97:319–25. [2] Silberbach M, Hannon D. Presentation of congenital heart disease in the neonate and young infant. Pediatr Rev/Am Acad Pediatr 2007;28:123–31. [3] Pina IL, O'Connor C. BNP-guided therapy for heart failure. JAMA 2009;301:432–4. [4] Troughton RW, Frampton CM, Yandle TG, et al. Treatment of heart failure guided by plasma aminoterminal brain natriuretic peptide (N-BNP) concentrations. Lancet 2000;355:1126–30. [5] Lainscak M, von Haehling S, Anker SD. Natriuretic peptides and other biomarkers in chronic heart failure: from BNP, NT-proBNP, and MR-proANP to routine biochemical markers. Int J Cardiol 2009;132:303–11. [6] El-Khuffash A, Molloy EJ. Are B-type natriuretic peptide (BNP) and N-terminal-proBNP useful in neonates? Arch Dis Child 2007;92:F320–4. [7] Mir TS, Falkenberg J, Friedrich B, et al. Levels of brain natriuretic peptide in children with right ventricular overload due to congenital cardiac disease. Cardiol Young 2005;15:396–401. [8] Kunii Y, Kamada M, Ohtsuki S, et al. Plasma brain natriuretic peptide and the evaluation of volume overload in infants and children with congenital heart disease. Acta Med Okayama 2003;57:191–7. [9] Koch A, Zink S, Singer H. B-type natriuretic peptide in paediatric patients with congenital heart disease. Eur Heart J 2006;27:861–6. [10] Czernik C, Lemmer J, Metze B, et al. B-type natriuretic peptide to predict ductus intervention in infants b 28 weeks. Pediatr Res 2008;64:286–90. [11] Cowley CG, Bradley JD, Shaddy RE. B-type natriuretic peptide levels in congenital heart disease. Pediatr Cardiol 2004;25:336–40. [12] Farombi-Oghuvbu I, Matthews T, Mayne PD, et al. N-terminal pro-B-type natriuretic peptide: a measure of significant patent ductus arteriosus. Arch Dis Child 2008;93: F257–60. [13] Sanjeev S, Pettersen M, Lua J, et al. Role of plasma B-type natriuretic peptide in screening for hemodynamically significant patent ductus arteriosus in preterm neonates. J Perinatol 2005;25:709–13. [14] Flynn PA, da Graca RL, Auld PA, et al. The use of a bedside assay for plasma B-type natriuretic peptide as a biomarker in the management of patent ductus arteriosus in premature neonates. J Pediatr 2005;147:38–42. [15] Choi BM, Lee KH, Eun BL, et al. Utility of rapid B-type natriuretic peptide assay for diagnosis of symptomatic patent ductus arteriosus in preterm infants. Pediatrics 2005;115:e255–61. [16] Chiruvolu A, Punjwani P, Ramaciotti C. Clinical and echocardiographic diagnosis of patent ductus arteriosus in premature neonates. Early Hum Dev 2009;85:147–9. [17] Yasue H, Yoshimura M, Sumida H, et al. Localization and mechanism of secretion of Btype natriuretic peptide in comparison with those of A-type natriuretic peptide in normal subjects and patients with heart failure. Circulation 1994;90:195–203. [18] Koch A, Singer H. Normal values of B type natriuretic peptide in infants, children, and adolescents. Heart (British Cardiac Society) 2003;89:875–8. [19] Suda K, Matsumura M, Matsumoto M. Clinical implication of plasma natriuretic peptides in children with ventricular septal defect. Pediatr Int 2003;45:249–54. [20] Law YM, Keller BB, Feingold BM, et al. Usefulness of plasma B-type natriuretic peptide to identify ventricular dysfunction in pediatric and adult patients with congenital heart disease. Am J Cardiol 2005;95:474–8. [21] Cannesson M, Bionda C, Gostoli B, et al. Time course and prognostic value of plasma B-type natriuretic peptide concentration in neonates undergoing the arterial switch operation. Anesth Analg 2007;104:1059–65 tables of contents. [22] Bolger AP, Sharma R, Li W, et al. Neurohormonal activation and the chronic heart failure syndrome in adults with congenital heart disease. Circulation 2002;106:92–9. [23] Davlouros PA, Kilner PJ, Hornung TS, et al. Right ventricular function in adults with repaired tetralogy of Fallot assessed with cardiovascular magnetic resonance imaging: detrimental role of right ventricular outflow aneurysms or akinesia and adverse right-to-left ventricular interaction. J Am Coll Cardiol 2002;40:2044–52. [24] Davlouros PA, Niwa K, Webb G, et al. The right ventricle in congenital heart disease. Heart (British Cardiac Society) 2006;92(Suppl 1):i27–38. [25] Coats AJ. Ethical authorship and publishing. Int J Cardiol 2009;131:149–50.
Contents lists available at ScienceDirect
International Journal of Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j c a r d
Diagnostic role of plasma BNP levels in neonates with signs of congenital heart disease Periklis A. Davlouros a,⁎, Ageliki A. Karatza b, Ioanna Xanthopoulou a, Gabriel Dimitriou b, Aikaterini Georgiopoulou a, Stefanos Mantagos b, Dimitrios Alexopoulos a a b
Department of Cardiology, Patras University Hospital, Greece Department of Paediatrics, Patras University Hospital, Greece
a r t i c l e
i n f o
Article history: Received 29 May 2009 Accepted 24 July 2009 Available online 26 August 2009 Keywords: Plasma BNP Congenital heart disease Neonates
a b s t r a c t Background: Only a few studies have examined the relationship of plasma BNP levels and congenital heart disease (CHD) in neonates and these mainly concern preterm neonates with patent ductus arteriosus. We aimed to investigate the diagnostic role of plasma BNP in neonates admitted in the neonatal intensive care unit, (NICU), with signs of congenital heart disease (CHD). Methods: Prospective assessment of plasma BNP levels in 75 consecutive neonates with suspected CHD (heart murmur, respiratory distress, or cyanosis), admitted in the NICU of our university hospital. The final diagnosis was done with echocardiography. Results: Haemodynamically significant Left to Right shunts, (hsLtR), were found in 29 neonates, insignificant LtR shunts in 22, no heart disease in 15 and cyanotic heart disease in 9. BNP levels were significantly higher in neonates with hsLtR shunts vs. all other groups (logBNP 2.9 ± 0.5 pg/ml vs. 1.5 ± 0.4 pg/ml vs. 1.5 ±0.3 pg/ ml vs. 1.6 ± 0.2 pg/ml, p b 0.0001). Plasma BNP levels N 132.5 pg/ml had 93.1% sensitivity and 100% specificity for diagnosing hsLtR shunts (accuracy 99.6%). Conclusions: Plasma BNP is a reliable test for diagnosing hsLtR shunts in the NICU. This will alert the neonatologist for ordering an echocardiographic examination, or if the latter is not available, for transferring the neonate to an appropriate tertiary centre with neonatal-paediatric cardiology facilities. Normal BNP levels imply the absence of a significant LtR shunt, but may not exclude cyanotic heart disease. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The overall prevalence of cardiovascular malformations at birth is approximately 4–6 per 1000 live births [1]. Nowadays many of them are diagnosed during intrauterine life, however a substantial proportion of neonates with congenital heart disease (CHD) are diagnosed after birth, either in neonatal intensive care units (NICU), or in and out of hospital basis after echocardiographic examination. The most frequent indications for an echocardiogram in the NICU are a heart murmur, respiratory distress and cyanosis [2]. However echocardiography is not always available in the NICU and when available, a properly trained sonographer in the diagnosis of CHD is also needed. Plasma brain natriuretic peptide (BNP), is a well known accurate and easy to conduct bedside laboratory test for the diagnosis and guidance of heart failure therapy in adults [3–5]. Unfortunately only a few studies have examined the relationship of plasma BNP levels and CHD
⁎ Corresponding author. Patras University Hospital, Rion, Patras, 26500, Greece. Tel.: +30 2610999281; fax: +30 6932468624. E-mail address: [email protected] (P.A. Davlouros). 0167-5273/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2009.07.029
in neonates and children [6–11]. It has been shown that plasma BNP levels are a useful adjunct to echocardiography in the diagnosis of patent ductus arteriosus and its response to treatment in premature neonates and in the diagnosis of persistent pulmonary hypertension of the newborn [6,10,12–15]. There is only one recent study showing that the use of a rapid bedside plasma BNP assay may be useful for the diagnosis of cardiovascular problems in neonates with respiratory distress during the first few days of their life in the NICU [10]. We aimed to investigate any possible diagnostic role of plasma BNP levels in neonates admitted in our NICU, with signs suggestive of heart disease. 2. Methods We examined 75 consecutive neonates which were admitted in our NICU, (Patras University Hospital), with signs suggestive of CHD. More specifically, a heart murmur, respiratory distress and/or cyanosis, were considered indices of possible CHD. The existence of the latter was ruled in or out with an echocardiogram. A haemodynamically significant Left to Right (hsLtR) shunt was defined as a Qp/Qs ≥ 1.5 in case of an atrial or ventricular septal defect (ASD, VSD). In case of a patent ductus arteriosus (PDA), a hsLtR shunt was defined as the existence of at least one of three clinical signs of severity (murmur, increased precordial activity, bounding pulses) and the coexistence of at least one of two echocardiographic criteria of severity, namely a left atrial/aortic root diameter (LA/Ao) N 1.5 and/or a LtR shunt with a pulsatile Doppler flow pattern at the level of the
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
43
Table 1 Echocardiographic diagnosis, presenting signs and plasma BNP levels (pg/ml) in neonates with hsLtR shunts. Echocardiographic diagnosis
Presenting sign
Plasma BNP levels (pg/ml)
Echocardiographic diagnosis
Presenting sign
Plasma BNP levels (pg/ml)
PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA PDA
Murmur RD Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur
5500 4920 4550 3798 3180 2360 2110 1830 1670 1220 651 290 277 116
PDA, ASD PDA, ASD PDA, ASD PDA, ASD PDA, ASD PDA, ASD, VSD ASD, VSD ASD, VSD ASD, VSDs ASD, VSD ASD, VSD ASD, VSD VSD VSD VSD
RD Murmur Murmur Murmur Murmur RD Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur
1290 1260 912 673 98.9 344 906 800 771 500 376 248 563 155 139
hsLtR shunt: haemodynamically significant Left to Right shunt. PDA: patent ductus arteriosus, RD: respiratory distress, ASD: atrial septal defect, and VSD: ventricular septal defect.
PDA [16]. Neonates with birth asphyxia, persistent pulmonary hypertension, or non congenital heart disease were excluded. Neonates were classified in 4 groups. Neonates with hsLtR shunts as defined above, neonates with haemodynamically insignificant LtR shunts (Qp/Qsb 1.5) and no significant PDA, neonates without heart disease and neonates with cyanotic heart disease. Plasma BNP levels were determined quantitatively in EDTAanticoagulated whole blood with a rapid bedside immunofluorometric assay kit (Triage BNP; Biosite Diagnostics, San Diego, CA) [15]. Blood samples were collected either before or after the echocardiogram and BNP measurement was conducted immediately. The exact timing of blood sampling is described in the Results section. Kolmogorov–Smirnov test showed that BNP levels, (pg/ml), were skewed and therefore results are expressed as medians (range) or were logarithmically transformed and expressed as mean ± SD. The Mann–Whitney U test and the Kruskal–Wallis test were used to assess differences of BNP levels between two or more than two groups respectively. The Student's t-test for independent samples and one-way ANOVA with Bonferroni post-hoc comparisons were used when BNP levels were logarithmically transformed. ROC curve analysis was used to examine the ability of BNP levels to detect significant PDA in premature neonates. All tests were two-tailed and statistical significance was considered for p-values b 0.05. All statistical analyses were performed using SPSS for windows (version 15.0 SPSS Inc., Chicago, IL, USA).
The study was approved by the ethics committee of our hospital and all parents gave informed consent for blood sampling of their neonates.
3. Results Of the 75 neonates studied, 34 (45.3%) were premature (b37 weeks gestational age, birth weight 1478.9 ± 653.2 g) with 10 of them b28 weeks gestational age (13.3% of total, birth weight 872 ± 112.1 g). Forty-one (54.7%) were normal term neonates (birth weight 3104 ± 410 g). The most frequent reason for conducting an echocardiogram in the NICU was the existence of a heart murmur (75%), followed by respiratory distress (13%) and cyanosis (12%). All but one neonate had a normal left ventricular (LV) fractional shortening and ejection fraction (EF), (35.3 ± 4.5% and 68.5 ± 6.3% respectively). Twenty-nine neonates (38.6%), had a hsLtR shunt, 22 (29.3%) had a haemodynamically insignificant LtR shunt, 15 (20%) did not have
Table 2 Echocardiographic diagnosis, presenting signs and plasma BNP levels (pg/ml) in neonates without hsLtR shunts. Echocardiographic diagnosis
Presenting sign
Haemodynamically insignificant LtR shunt ASD Murmur ASD Murmur ASD Murmur ASD Murmur ASD Murmur ASD Murmur ASD Arrhythmia ASD, dextrocardia Murmur VSD Murmur VSD Murmur VSD Murmur VSD, PFO Murmur VSD, PFO Murmur VSD, ASD Murmur VSD, ASD Murmur VSD, ASD Murmur VSD, ASD Murmur PDA Murmur PDA Murmur PDA Murmur PDA RD PDA, ASD RD
BNP levels (pg/ml) 12.5 5 14.4 27.6 100 45 48 23.5 48.4 71.1 45.8 75 25.4 6.5 15.9 7.5 126 25.6 93.5 46 65 82.2
Echocardiographic diagnosis Cyanotic heart disease DORV, VSD TGA, ASD TGA, VSD TGA, VSD TAPVR, PDA TAPVR Critical PS, ASD PA-VSD PA-VSD No heart disease Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal
Presenting signs
BNP levels (pg/ml)
Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis Cyanosis
95 35.9 26.4 35 45 39.1 29.8 54.5 120
RD RD RD RD RD Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur Murmur
56.9 55 74.7 29.3 10.1 11.3 36.2 14 83 8.9 16.6 44.1 42 50.2 14.8
PDA: patent ductus arteriosus, RD: respiratory distress, ASD: atrial septal defect, VSD: ventricular septal defect, DORV: double outlet right ventricle, TGA: transposition of the great arteries, TAPVR: total anomalous venous return, PS: pulmonary stenosis, and PA: pulmonary atresia.
44
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
Fig. 1. Plasma BNP levels in relation to the underlying echocardiographic diagnosis. From 75 consecutive neonates with signs of heart disease, neonates with hsLtR shunts had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (ANOVA p b 0.0001). (hsLtR shunt: haemodynamically significant Left to Right shunt and CHD: congenital heart disease).
heart disease and 9 (12%) had cyanotic heart disease (Tables 1 and 2). Blood sampling was done within 12 h from birth in 9 neonates with cyanosis and in one neonate with heart failure signs (27 weeks gestational age with a massive PDA, EF 40%). Blood sampling in the remaining neonates was done beyond the second day of life. 3.1. BNP levels in neonates with hsLtR shunts Plasma BNP levels were significantly higher in neonates with hsLtR shunts compared to the other three groups (median BNP 800 pg/ml vs. 45.4 pg/ml vs. 36.2 pg/ml vs. 39.1 pg/ml p b 0.001, logBNP 2.9 ± 0.5 pg/ ml vs. 1.5 ± 0.4 pg/ml vs. 1.5 ± 0.3 pg/ml vs. 1.6 ± 0.2 pg/ml ANOVA
Fig. 3. Plasma BNP levels in relation to the underlying echocardiographic diagnosis. From 75 consecutive neonates with signs of heart disease, neonates with hsPDA had significantly higher plasma BNP levels compared to neonates with hsLtR shunts other than PDA (p = 0.016). Both groups had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (ANOVA p b 0.0001). (hsPDA: haemodynamically significant patent ductus arteriosus, hsLtR shunt: haemodynamically significant Left to Right shunt, and CHD: congenital heart disease).
p b 0.0001, Bonferroni p b 0.0001), (Fig. 1). ROC curve analysis showed that plasma BNP levelsN 132.5 pg/ml had 93.1% sensitivity and 100% specificity for diagnosing hsLtR shunts and BNP levels b 96.9 pg/ml had 100% sensitivity and 93.5% specificity for ruling out hsLtR shunts. The overall test accuracy was 99.6% (Fig. 2). When excluding extremely premature neonates (b28 weeks, N = 10) from analysis, the results did not change. Plasma BNP levels did not correlate significantly with the EF in the total population of the study. When examining only neonates with hsLtR shunts, there was a marginal negative correlation between BNP and EF (r = −0.47, p = 0.05). 3.2. BNP levels in neonates with hsPDA Neonates with hsPDA (N = 19) had significantly higher plasma BNP levels compared to the rest of neonates with hsLtR shunts (N = 10), (logBNP 3.06 ± 0.5 pg/ml vs. 2.6 ± 0.3 pg/ml p = 0.016). Neonates with hsPDA also had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (logBNP 3.06 ± 0.5 pg/ml vs. logBNP 1.5 ± 0.4 pg/ml vs. 1.5 ± 0.3 pg/ml vs. 1.6 ± 0.2 pg/ml ANOVA p b 0.001, Bonferroni p b 0.0001), (Fig. 3). ROC analysis showed that a BNP N 201.5 pg/ml had 89.5% sensitivity and 100% specificity for diagnosing a hsPDA and a BNPb 96.9 pg/ml had 100% sensitivity and 93.5% specificity for ruling out a hsPDA. 3.3. BNP levels in neonates with hsLtR shunts other than PDA
Fig. 2. ROC analysis. Plasma BNP levels N 132.5 pg/ml had 93.1% sensitivity and 100% specificity for diagnosing hsLtR shunts. The overall test accuracy, (area under the curve), was 99.6%. (hsLtR shunt: haemodynamically significant Left to Right shunt).
Neonates with hsLtR shunts other than PDA, (i.e. ASD and VSD, N = 10), had significantly higher plasma BNP compared to neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease (logBNP 2.6 ± 0.3 pg/ml vs. logBNP 1.5 ± 0.4 pg/ml vs. 1.5 ± 0.3 pg/ml vs. 1.6 ± 0.2 pg/ml ANOVA p b 0.001, Bonferroni p b 0.0001), (Fig. 3). A BNP of 147 pg/ml had 100% sensitivity and specificity in this group.
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
3.4. BNP levels in premature neonates with hsLtR shunts When examining only premature neonates (N = 34), 20 of them (59%) had a hsLtR shunt, most frequently PDA (N = 16). The latter had significantly higher BNP levels compared to neonates with insignificant shunts and neonates without heart disease (logBNP 2.9 ± 0.56 pg/ml vs. logBNP 1.4 ± 0.4 pg/ml vs. 1.3 ± 0.3 pg/ml ANOVA p b 0.001, Bonferroni p b 0.0001). In premature neonates with a significant PDA a BNP N 283.5 pg/ml had 81.3% sensitivity and 100% specificity for diagnosing a significant PDA and a BNP of 193.5 pg/ml had 85% sensitivity and 100% specificity for diagnosing any hsLtR shunt. Ten premature neonates had a gestational age b 28 weeks. From these, 2 had an insignificant PDA (BNP 65 and 93.5 pg/ml) and 8 had hsPDA on echocardiographic examination. In the latter, BNP ranged from 116 to 5500 pg/ml (median 2814 pg/ml) and all of them needed intervention for duct closure, mainly indomethacin therapy. 4. Discussion B-type natriuretic peptide (BNP) is a hormone secreted by the ventricles under haemodynamic stress associated with ventricular enlargement with or without ventricular systolic dysfunction [17]. This study aimed to investigate the diagnostic accuracy of plasma BNP levels in neonates with suspected CHD. Echocardiography is the main diagnostic tool in this setting, however the echocardiographic equipment or a properly trained sonographer in the diagnosis of CHD may not be available in the NICU. In the present study we showed that bedside rapid BNP assessment may aid in the diagnosis or exclusion of significant LtR shunts with volume loading. However, it may not be helpful in cases of cyanotic CHD. 4.1. BNP levels in neonates with hsLtR shunts Plasma BNP levels are elevated in healthy newborns right after birth and then rapidly decrease to b100 pg/ml on day two and b50 pg/ml beyond the fourth day of life [18]. This may reflect volume loading as a result of the transition from fetal to neonatal circulation, or diminished clearance of the peptide from the placenta. We measured plasma BNP levels after the second day of life, except in neonates with cyanosis in whom a rapid and prompt diagnosis is mandatory. We found that overall neonates with a volume loading LtR shunt have significantly elevated plasma BNP levels compared to neonates with haemodynamically insignificant LtR shunts. The latter had BNP levels comparable to neonates without evidence of CHD on echocardiography. The observation that plasma BNP levels are elevated in neonates with hsLtR shunts compared to the rest in our study probably reflects the mechanism of BNP secretion by the ventricles when they are under haemodynamic stress i.e. released by the ventricular myocytes in response to volume and possibly pressure load. Our results are in accord with the few studies which have examined the relationship of plasma BNP and LtR shunting, mainly in children. Plasma BNP levels have been found to be elevated in children with LV and RV volume overload (ASD, VSD, and PDA) without ventricular dysfunction and to correlate significantly with Qp/Qs and LV and RV end-diastolic volumes [8,19]. It has also been shown that in children with LtR shunts plasma BNP levels correlate with shunt volume and systolic RV pressure [9]. Elevated plasma BNP levels have been found in children with CHD and reduced LV systolic function [20]. A negative correlation between plasma BNP levels and LV systolic function has also been described in paediatric patients with CHD in a study including patients with impaired LV systolic function [9]. We also found a weak negative correlation between LV ejection fraction and plasma BNP levels in neonates with hsLtR shunts, extending the above findings to the neonatal population with CHD. However, as expected, the LV function was normal in our population, as LtR shunts in paediatric patients initially lead to right or left ventricular volume overload prior to the emergence of ventricular dysfunction [8]. We conclude that in
45
neonates with hsLtR shunts, plasma BNP levels may reflect the degree of ventricular distension-stretching and secondarily ventricular systolic impairment, (if any), as a result of volume load. This implies that plasma BNP levels may be a useful index of the severity of volume loading in neonates with LtR shunts and may aid in determining the need for an intervention. However, the latter was not examined in our study. There are not many studies examining the diagnostic role of plasma BNP in neonates with suspected CHD. One recent study showed that a bedside plasma BNP assay may be useful for the diagnosis of cardiovascular problems in neonates with respiratory difficulties during the first few days of their life [10]. In this study, plasma BNP levels were significantly higher in neonates with cardiovascular disease compared to the rest with cutoff points of 346, 421, 570.5 and 191.5 pg/pl within 18 h, at 18–36 h, at 36–60 h and beyond 60 h of life. In our study blood sampling was done beyond 48 h for all neonates apart from those with cyanosis and the overall cutoff value was 132.5 pg/ml, 201.5 pg/ml for a hsPDA and 147 pg/ml for hsLtR shunts other than PDA. Our study differs however from that of Ko et al. in that all neonates with suspected heart disease were included and not only those with respiratory problems. 4.2. BNP levels in term and preterm neonates with hsPDA or other hsLtR shunts In contrast to the whole population of neonates admitted in NICUs, there is enough evidence that plasma BNP levels are a useful adjunct to echocardiography in the diagnosis of hsPDA in preterm infants [13,15]. In neonates b28 weeks gestational age a BNP of N300 pg/ml beyond the second day of life, predicts a hsPDA [14]. It has also been shown that a BNP N 550 pg/ml on the second day of life in such neonates predicts interventions for ductus closure (sensitivity 83%, specificity 86%). Successful closure is reflected by a corresponding decrease in BNP and at a cutoff of 70 pg/ml, BNP is a useful screening tool for diagnosis and for monitoring efficacy of treatment of hsPDA [13]. Our study included 10 neonates with a gestational age b 28 weeks and all of them had a PDA on echocardiography. In 8 of them the PDA was haemodynamically significant, all of them needed intervention for ductus closure, mainly indomethacin treatment, and BNP ranged from 116 to 5500 pg/ml, (median 2814 pg/ml), in accord with the above mentioned studies. In the total population examined, neonates with hsPDA had significantly higher plasma BNP levels compared to neonates with hsLtR shunts, neonates with insignificant shunts, neonates without heart disease and neonates with cyanotic heart disease. Neonates with hsLtR shunts other than PDA, (i.e. ASD and VSD), had significantly higher plasma BNP compared to the latter three groups (Fig. 3). We might conclude that elevated plasma BNP levels in term or preterm neonates admitted in the NICU with signs suggestive of CHD, imply the existence of a significant volume loading LtR shunt. Preterm infants with a hsPDA have the highest BNP levels, especially those b28 weeks gestational age. The cutoff values of plasma BNP levels for the detection of a hsLtR shunt other than a PDA may be lower. 4.3. BNP levels in neonates with cyanotic CHD Nine neonates (12%), had cyanotic CHD. These neonates did not have elevated plasma BNP levels compared to neonates with insignificant LtR shunts or without heart disease, even though in this group BNP was determined within 48 h from birth, at a time when levels are expected to be relatively elevated [18]. It has been shown that in neonates with transposition of the great arteries (TGA), plasma BNP levels before arterial switch operation are not elevated [21]. On the contrary BNP levels increase beyond 6 h post switch operation in neonates with haemodynamic compromise [21]. In infants and children with significant pulmonary valve stenosis or unoperated tetralogy of Fallot, plasma BNP levels are not elevated compared to healthy controls [9]. Cyanosis itself does not seem to affect plasma BNP levels, as there is no correlation between the latter and oxygen saturation [11]. In another
46
P.A. Davlouros et al. / International Journal of Cardiology 147 (2011) 42–46
study of a very heterogeneous group of children with different types of CHD, there was no correlation between plasma BNP levels and RV outflow obstruction [11]. Data on patients with severe obstruction of the right or left heart demonstrate that not all kinds of increasing ventricular pressure cause directly an increase of plasma BNP, implying that the mechanism of BNP release may be much more complex in CHD [9]. Our study demonstrated a relationship between BNP levels and hsLtR shunts; however a simple correlation between plasma BNP levels and pressure or volume load regardless the underlying specific heart defect may not exist. When considering more heterogeneous populations of CHD patients elevated plasma BNP levels may reflect more accurately the degree of ventricular impairment [9]. Therefore in neonates with cyanotic heart disease without ventricular dysfunction, the haemodynamic stress on the left or right ventricle may not be severe enough for stimulating BNP elevation. We speculate that in these neonates, low plasma BNP levels reflect a compensated status of the heart and cannot preclude any pathology. On the contrary, in adults with repaired cyanotic CHD and heart failure (e.g. tetralogy of Fallot), plasma BNP levels are elevated [22]. Right ventricular dilatation and dysfunction due to pulmonary regurgitation has been shown in these patients and this kind of decompensation may explain the increase in BNP levels [23,24]. Further research is needed on the relationship of cyanotic CHD and BNP levels in neonates. 5. Conclusion Bedside rapid plasma BNP assay is a reliable, relatively accurate and easy to conduct blood test, for diagnosing hsLtR shunts in neonates admitted to the NICU with symptoms and/or signs of suspected CHD. This will alert the neonatologist for ordering an echocardiographic examination, or if the latter is not available, for transferring the neonate to an appropriate tertiary centre with neonatal-paediatric cardiology facilities. When plasma BNP levels are not elevated, there is a certain level of assurance that a haemodynamically significant LtR shunt does not exist, however major cyanotic CHD cannot be excluded. 6. Limitations and strengths of the study Although the number of neonates with significant CHD is adequate in this study, most hsLtR shunts represented PDA. More work is needed to definitively conclude about the role of plasma BNP levels in neonates with other hsLtR shunts like ASD and VSD, as well as its predictive value regarding the need for intervention. However, our study aimed to determine the adjunctive role of BNP in diagnosing CHD in the NICU and in this context we demonstrate some useful information. It is important that this study included only neonates and not a mixed population of infants, children or adults with CHD like all the other studies in the literature. Although our results were in accordance with the above studies to a certain degree, therefore extending their conclusions to the neonatal population, we should stress that data regarding older patients cannot be transferred directly to neonates and BNP cutoff values may be different between neonates and older patients. We consider the results regarding cyanotic CHD very interesting. Certainly the numbers are small, but hypothesis generating and stimulating further research in the field. Finally, haemodynamic data would be desirable in such studies, however this was not possible in our centre.
Acknowledgment The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [25]. References [1] Loffredo CA. Epidemiology of cardiovascular malformations: prevalence and risk factors. Am J Med Genet 2000;97:319–25. [2] Silberbach M, Hannon D. Presentation of congenital heart disease in the neonate and young infant. Pediatr Rev/Am Acad Pediatr 2007;28:123–31. [3] Pina IL, O'Connor C. BNP-guided therapy for heart failure. JAMA 2009;301:432–4. [4] Troughton RW, Frampton CM, Yandle TG, et al. Treatment of heart failure guided by plasma aminoterminal brain natriuretic peptide (N-BNP) concentrations. Lancet 2000;355:1126–30. [5] Lainscak M, von Haehling S, Anker SD. Natriuretic peptides and other biomarkers in chronic heart failure: from BNP, NT-proBNP, and MR-proANP to routine biochemical markers. Int J Cardiol 2009;132:303–11. [6] El-Khuffash A, Molloy EJ. Are B-type natriuretic peptide (BNP) and N-terminal-proBNP useful in neonates? Arch Dis Child 2007;92:F320–4. [7] Mir TS, Falkenberg J, Friedrich B, et al. Levels of brain natriuretic peptide in children with right ventricular overload due to congenital cardiac disease. Cardiol Young 2005;15:396–401. [8] Kunii Y, Kamada M, Ohtsuki S, et al. Plasma brain natriuretic peptide and the evaluation of volume overload in infants and children with congenital heart disease. Acta Med Okayama 2003;57:191–7. [9] Koch A, Zink S, Singer H. B-type natriuretic peptide in paediatric patients with congenital heart disease. Eur Heart J 2006;27:861–6. [10] Czernik C, Lemmer J, Metze B, et al. B-type natriuretic peptide to predict ductus intervention in infants b 28 weeks. Pediatr Res 2008;64:286–90. [11] Cowley CG, Bradley JD, Shaddy RE. B-type natriuretic peptide levels in congenital heart disease. Pediatr Cardiol 2004;25:336–40. [12] Farombi-Oghuvbu I, Matthews T, Mayne PD, et al. N-terminal pro-B-type natriuretic peptide: a measure of significant patent ductus arteriosus. Arch Dis Child 2008;93: F257–60. [13] Sanjeev S, Pettersen M, Lua J, et al. Role of plasma B-type natriuretic peptide in screening for hemodynamically significant patent ductus arteriosus in preterm neonates. J Perinatol 2005;25:709–13. [14] Flynn PA, da Graca RL, Auld PA, et al. The use of a bedside assay for plasma B-type natriuretic peptide as a biomarker in the management of patent ductus arteriosus in premature neonates. J Pediatr 2005;147:38–42. [15] Choi BM, Lee KH, Eun BL, et al. Utility of rapid B-type natriuretic peptide assay for diagnosis of symptomatic patent ductus arteriosus in preterm infants. Pediatrics 2005;115:e255–61. [16] Chiruvolu A, Punjwani P, Ramaciotti C. Clinical and echocardiographic diagnosis of patent ductus arteriosus in premature neonates. Early Hum Dev 2009;85:147–9. [17] Yasue H, Yoshimura M, Sumida H, et al. Localization and mechanism of secretion of Btype natriuretic peptide in comparison with those of A-type natriuretic peptide in normal subjects and patients with heart failure. Circulation 1994;90:195–203. [18] Koch A, Singer H. Normal values of B type natriuretic peptide in infants, children, and adolescents. Heart (British Cardiac Society) 2003;89:875–8. [19] Suda K, Matsumura M, Matsumoto M. Clinical implication of plasma natriuretic peptides in children with ventricular septal defect. Pediatr Int 2003;45:249–54. [20] Law YM, Keller BB, Feingold BM, et al. Usefulness of plasma B-type natriuretic peptide to identify ventricular dysfunction in pediatric and adult patients with congenital heart disease. Am J Cardiol 2005;95:474–8. [21] Cannesson M, Bionda C, Gostoli B, et al. Time course and prognostic value of plasma B-type natriuretic peptide concentration in neonates undergoing the arterial switch operation. Anesth Analg 2007;104:1059–65 tables of contents. [22] Bolger AP, Sharma R, Li W, et al. Neurohormonal activation and the chronic heart failure syndrome in adults with congenital heart disease. Circulation 2002;106:92–9. [23] Davlouros PA, Kilner PJ, Hornung TS, et al. Right ventricular function in adults with repaired tetralogy of Fallot assessed with cardiovascular magnetic resonance imaging: detrimental role of right ventricular outflow aneurysms or akinesia and adverse right-to-left ventricular interaction. J Am Coll Cardiol 2002;40:2044–52. [24] Davlouros PA, Niwa K, Webb G, et al. The right ventricle in congenital heart disease. Heart (British Cardiac Society) 2006;92(Suppl 1):i27–38. [25] Coats AJ. Ethical authorship and publishing. Int J Cardiol 2009;131:149–50.