Pediatr Cardiol DOI 10.1007/s00246-014-1075-3
ORIGINAL ARTICLE
Clinical Utility of the Plasma Brain Natriuretic Peptide Level in Monitoring Tetralogy of Fallot Patients over the Long Term After Initial Intracardiac Repair: Considerations for Pulmonary Valve Replacement Atsushi Kitagawa • Norihiko Oka • Sumito Kimura • Hisashi Ando • Takashi Honda • Manabu Takanashi • Eri Mineo • Kagami Miyaji • Masahiro Ishii Received: 16 August 2014 / Accepted: 1 December 2014 Ó Springer Science+Business Media New York 2014
Abstract Clinicians are currently encountering an increasing number of patients in the long-term period after tetralogy of Fallot (TOF) repair presenting with pulmonary valve regurgitation (PR) or right ventricular (RV) dysfunction. The purpose of this study was to evaluate the clinical utility of the plasma brain natriuretic peptide (BNP) level and consider surgical indications and timing of pulmonary valve replacement (PVR). We examined 33 patients (21 males, 12 females, mean age 14.5 ± 2.8 years) who underwent TOF repair at Kitasato University Hospital. All patients were evaluated using echocardiography and blood sampling. The mean age at the time of initial repair was 1.3 ± 0.7 years. The patients with moderate–severe PR exhibited significantly higher plasma BNP levels than the patients with trivial–mild PR (mean 37.5 ± 33.1 vs. 17.3 ± 6.6 pg/ml, p = 0.013). The mean plasma BNP level with cardiac symptoms was higher than that observed in the patients without any symptoms (71.4 ± 46.1 vs. 25.0 ± 14.0 pg/ml, p = 0.005). The mean BNP level was significantly decreased after PVR (71.3 ± 46.1–26.1 ± 13.2 pg/ml, p = 0.009), and the plasma BNP level was found to be positively correlated with the RV end-diastolic pressure (r = 0.851; p = 0.008). The optimal BNP cut-off value for considering PVR was 32.15 pg/ml (sensitivity, 85.7 %; specificity, 83.3 %). The plasma BNP level may become a
A. Kitagawa S. Kimura H. Ando T. Honda M. Takanashi E. Mineo M. Ishii Department of Pediatrics, Kitasato University School of Medicine, Sagamihara, Japan N. Oka (&) K. Miyaji Department of Cardiovascular Surgery, Kitasato University School of Medicine, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0374, Japan e-mail: [email protected]
useful diagnostic tool for considering the indications and optimal timing of PVR over the long term after TOF repair. Keywords Tetralogy of Fallot Pulmonary valve regurgitation Pulmonary valve replacement Brain natriuretic peptide Abbreviations BNP Brain natriuretic peptide CHD Congenital heart disease CMR Cardiovascular magnetic resonance imaging EDP End-diastolic pressure EDVI End-diastolic volume index EF Ejection fraction ESVI End-systolic volume index PR Pulmonary valve regurgitation PVR Pulmonary valve replacement RVOTR Right ventricular outflow tract reconstruction TOF Tetralogy of Fallot
Introduction Since the first intracardiac surgery for tetralogy of Fallot (TOF) was reported by Lillehei et al. in 1954 [14], the outcomes of corrective surgery have improved. Clinicians are now encountering an increasing number of patients suffering from late surgical complications, such as heart failure, arrhythmia, and sudden death [21]. For example, Murphy et al. reported that the long-term survival rate of patients treated with TOF repair is lower than that observed in the general population [16]. Pulmonary valve regurgitation (PR) is the primary cause of abnormal hemodynamics over the long term after repair of
123
Pediatr Cardiol
TOF. Severe PR contributes to the development of chronic right ventricular (RV) volume overload and can lead to RV dilatation, RV dysfunction, life-threatening ventricular arrhythmia, and sudden death [6, 7]. There is strong evidence that pulmonary valve replacement (PVR) is highly effective in eliminating or greatly reducing PR [5, 19]. In addition, there is a general consensus that patients exhibiting symptoms related to severe PR and RV dysfunction are candidates for PVR and that physicians should treat these patients before RV dysfunction becomes irreversible. In fact, waiting for symptoms to develop may make it too late to provide adequate treatment. Several studies recommend determining the indications for PVR based on assessments using echocardiography, cardiac catheterization, and cardiovascular magnetic resonance imaging (CMR). However, there are various technical, economic, and invasive problems associated with these diagnostic tools. In recent decades, the level of plasma brain natriuretic peptide (BNP) has received significant attention in the field of heart failure as a marker in the general population [25]. In addition, it is known that the plasma BNP levels are elevated in asymptomatic or only mildly symptomatic patients with chronic RV pressure overload or volume expansion [15, 17], and Eindhoven et al. reported the usefulness of BNP in monitoring patients with complex congenital heart diseases (CHDs), such as TOF [4]. However, the clinical utility of measuring the BNP level to determine the indication for PVR after TOF repair remains controversial. The purpose of this study was, therefore, to evaluate the plasma BNP levels over the long term in patients with a history of TOF repair and identify the indications and optimal surgical timing for PVR.
Patients and Methods Patients This single-institutional study was undertaken at Kitasato University Hospital. Patients older than 10 years of age who had been treated with intracardiac TOF repair were studied.
sampling, including measurement of the plasma BNP level, once or twice a year. In this study, blood was collected in EDTA and lithium heparin vacutainer tubes and immediately sent to the laboratory for an analysis, where the BNP level was assayed. The plasma BNP level was determined using a commercially available chemiluminescence enzyme immunoassay (CLEIA) (MI02 Shionogi BNP kit; Shionogi, Osaka, Japan). The upper normal limit of BNP was 18.4 pg/ml. The severity of PR was assessed based on the width of the vena contracta measured using color Doppler echocardiography. From the color-flow Doppler across the pulmonary valve, the regurgitant jet width was measured at a valve level during early diastole. The pulmonary valve diameter was also measured. The ratio of regurgitant jet width to pulmonary valve diameter was then calculated. We defined moderate–severe PR as PR width/ pulmonary valve ratio C0.5. And we also defined trivial– mild PR as PR width/pulmonary valve ratio \0.5. Cardiac catheterization studies were performed in all seven patients prior to PVR and at 1 year after PVR. We measured both ventricular end-systolic and end-diastolic volumes, and ejection fraction using cardiac catheterization. Statistical Analysis The data are expressed as the median (range) or mean ± standard deviation (SD), as appropriate. Associations between two continuous variables were assessed using a linear regression analysis. Comparisons between paired groups were made using the paired t test or Wilcoxon signed-rank test, as appropriate, while comparisons between independent groups were made using the Mann– Whitney test. A receiver operating characteristic (ROC) curve analysis was performed to determine the BNP cut-off value for predicting symptoms associated with RV dysfunction. The SPSS version 15.0 (SPSS Inc., Chicago, Illinois) software program was used for the statistical analysis, and a p value of \0.05 was considered to be statistically significant.
Results Indication for PVR Clinical Characteristics of the Patients The indications for PVR at our institution included any signs or symptoms attributable to RV dilatation and cardiac dysfunction, such as general fatigue, dyspnea, palpitation and exercise intolerance. Cardiac Studies At our institution, all TOF patients undergo electrocardiography, chest X-ray, echocardiography, and blood
123
Total 33 patients were enrolled. The mean age at the initial intracardiac repair procedure was 1.3 ± 0.7 years, and the mean interval between repair and the examination was 12.8 ± 2.6 years (Table 1). Twenty-seven patients were found to have moderate–severe PR. Six patients were found to have trivial–mild PR. Seven patients underwent PVR due to symptoms associated with cardiac dysfunction (Fig. 1). There were no significant differences between the
Pediatr Cardiol Table 1 Patient characteristics Variable
Table 2 Patient characteristics and severity of PR N = 33
Age at examination (years)
14.5 ± 2.8
Sex (male:female)
21:12
Age at initial repair (years)
1.3 ± 0.7
Interval between initial repair and examination (years)
12.8 ± 2.6
The values are presented as the mean ± standard deviation (SD)
Variable
Moderate– severe PR (n = 27)
Trivial– mild PR (n = 6)
p value
Age at initial repair (years)
1.3 ± 0.7
1.4 ± 0.8
0.310
Age at examination (years)
12.4 ± 5.9
12.9 ± 1.0
0.247
Interval between repair and examination (years)
10.6 ± 5.5
11.5 ± 1.1
0.441
Transannular patch
20
4
Pulmonary valvotomy
4
2
Monocusp valve
3
–
Trivial–mild
21
6
Moderate–severe
6
–
I
23
6
II
3
–
III
1
–
Type of RVOTR
TR grade
NYHA functional class
PR pulmonary valve regurgitation, RVOTR right ventricular outflow tract reconstruction, TR tricuspid valve regurgitation, NYHA New York Heart Association
Fig. 1 Therapeutic strategies employed at our institution PR pulmonary valve regurgitation, PVR pulmonary valve replacement
patients with moderate–severe PR and those with trivial– mild in terms of the mean age at initial repair, mean age at the examination, or interval from repair to the examination. The types of RV outflow tract reconstruction (RVOTR) used in the initial operation are shown in Table 2. Twentynine patients were classified as having a New York Heart Association (NYHA) functional class I status, and four patients were classified as having NYHA functional class II or III status. Relationship Between the Plasma BNP Level and the Severity of PR The mean plasma BNP level in the patients with moderate– severe PR was significantly higher than that observed in the patients with trivial–mild PR (37.5 ± 33.1 vs. 17.3 ± 6.6 pg/ml; p = 0.013) (Fig. 2a). The mean plasma BNP level of patients with symptoms associated with cardiac dysfunction was significantly higher than that observed in the patients having moderate–severe PR without any symptoms (71.4 ± 46.1 vs. 25.0 ± 14.0 pg/ml; p = 0.005) (Fig. 2b).
Fig. 2 Associations between the plasma BNP level and the PR severity and symptoms. a Association between the plasma BNP level and the severity of PR. b Comparison of the plasma BNP levels of moderate–severe PR patients with cardiac symptoms and without
Relationships Between the Plasma BNP Levels and Cardiac Catheterization Parameters Obtained Before PVR Cardiac catheterization was performed in all seven patients prior to PVR. The plasma BNP level before PVR was found to be positively correlated with the RV end-diastolic pressure (EDP) (r = 0.851; p = 0.008), LV EDP (r = 0.755; p = 0.025), and LV end-diastolic volume
123
Pediatr Cardiol
index (EDVI) (r = 0.672; p = 0.049) (Fig. 3b, c, d). Meanwhile, the BNP level was found to be negatively correlated with the RV EDVI (r = -0.680; p = 0.046) (Fig. 3a). In contrast, the RV end-systolic volume index (ESVI), RV ejection fraction (EF), and left ventricular (LV) EF were not found to be correlated with the plasma BNP level (Fig. 3e, f, g, and h). The BNP Cut-Off Value for Considering PVR Figure 4 shows the ROC curve for the plasma BNP level in predicting symptoms associated with RV dysfunction. The area under the curve (AUC) for the plasma BNP level was 0.893, and the optimal cut-off value for the plasma BNP Fig. 3 Relationships between the plasma BNP level the and cardiac catheterization parameters
123
level for considering PVR was 32.15 pg/ml (sensitivity, 85.7 %; specificity, 83.3 %).
Changes One Year After PVR The mean plasma BNP level 1 year after PVR was significantly lower than that observed before PVR (26.1 ± 13.2 vs. 71.4 ± 46.1 pg/ml; p = 0.009) (Fig. 5a). In addition, the QRS duration after PVR was significantly narrower than that measured before PVR (mean 152 ± 12 vs. 167 ± 16 ms; p = 0.009) (Fig. 5b). Among the cardiac catheterization parameters, the mean RV EDVI 1 year after PVR was significantly smaller than that observed before
Pediatr Cardiol
PVR (106.9 ± 14.7 vs. 126.0 ± 19.5 ml/m2; p = 0.008) (Table 3). In all seven patients, the clinical symptoms significantly improved after PVR (Fig. 5c).
Discussion The early and mid-term outcomes of TOF repair have improved in recent years [18]. As the number of long-term survivors has increased, a growing number of late surgical complications have been reported. Numerous studies have shown that patients with a history of TOF repair exhibit many risk factors for exercise intolerance, heart failure, arrhythmia, and sudden death [2, 7, 11, 16, 21, 22]. PR is the primary cause of abnormal hemodynamics after TOF repair, particularly in patients requiring a transannular patch during the procedure. In the present study, our data demonstrated that 27 patients (81.8 %) had moderate– severe PR at a mean of 12.8 ± 2.6 years after the initial Table 3 Changes in the cardiac catheterization parameters before and after PVR Variable
Before PVR
After PVR
p value
2
126.0 ± 19.5
106.9 ± 14.7
0.008
2
RV ESVI (ml/m )
58.2 ± 8.2
52.9 ± 11.6
0.088
RV EDP (mmHg)
6.7 ± 3.3
8.3 ± 4.7
0.058 0.129
RV EDVI (ml/m )
53.1 ± 8.2
50.5 ± 9.0
LV EDVI (ml/m2)
RV EF (%)
105.4 ± 19.6
109.4 ± 21.1
0.199
LV ESVI (ml/m2)
45.3 ± 12.3
45.0 ± 13.8
0.161
LV EDP (mmHg) LV EF (%) Fig. 4 Receiver operating characteristic (ROC) curve analysis of the plasma BNP levels for predicting symptoms associated with RV dysfunction
7.9 ± 1.6
11.2 ± 3.0
0.015
57.0 ± 9.1
59.7 ± 6.0
0.252
RV right ventricle, EDVI end-diastolic volume index, ESVI end-systolic volume index, EDP end-diastolic pressure, EF ejection fraction, LV left ventricle
Fig. 5 Changes in the parameters before and after PVR. a Plasma BNP level. b QRS duration. c NYHA functional class
123
Pediatr Cardiol
intracardiac repair procedure. Severe PR contributes to the development of RV pressure overload and RV dilatation, leading to RV dysfunction. Furthermore, RV dilatation exhibits a close association with prolongation of the QRS duration [1, 3]. PVR is a significantly effective procedure for eliminating or reducing PR [5, 19]. Clinically, several studies have reported significant improvements in the NYHA functional class after PVR [5, 8–10, 20, 24]. In our series, seven patients were treated with PVR. In all of these seven patients, the PR disappeared in association with improvements in clinical symptoms. It is well established that patients with symptoms related to severe PR and RV dysfunction are candidates for PVR. At our institution, we had performed PVR in symptomatic patients, such as general fatigue, palpitation, and exercise intolerance. However, the presence of RV volume overload or dysfunction due to severe PR does not necessarily induce clinical symptoms. Severe PR can be tolerated without any symptoms for many years, while the underlying RV dysfunction becomes irreversible. In fact, waiting for symptoms to develop may make it too late to determine the timing of PVR. Therefore, PVR should be considered before RV dysfunction becomes irreversible. Several studies have suggested the optimal surgical indications for PVR. For example, CMR is a popular modality for measuring the ventricular volume, cardiac function, and valve regurgitation fractions after TOF repair. On the other hand, CMR takes much cost than other examinations. Furthermore, CMR requires a lot of time to measure cardiac parameters. Numerous investigators have attempted to measure the RV function or severity of PR using echocardiography. However, in the absence of significant echocardiographic markers, assessing the RV volume and function after TOF repair remains challenging. The development of other easy, reliable, costless diagnostic tool for detecting cardiac dysfunction in asymptomatic patients with chronic PR is desirable. The plasma BNP level has received significant interest in recent decades. BNP is synthesized and released into the circulation by ventricular myocytes in response to pressure overload, volume expansion, and increased myocardial wall stress. This biomarker exhibits clinical utility for assessing the degree of cardiac impairment. BNP is known to be a well-established marker of heart failure in the general population [23]. The plasma BNP levels are also increased in patients with complex CHD (including TOF, systemic RV, and univentricular hearts) compared with the reference values and/or levels observed in controls, even among asymptomatic patients [4]. The present data showed the plasma BNP levels to be significantly higher in the patients with moderate–severe PR than in those with trivial–mild PR. Among cardiac catheterization parameters, higher RV EDP, LV EDP, and
123
LV EDVI values were found to be associated with a higher BNP level. Severe PR contributes to the development of symptoms associated with RV pressure overload, and biventricular pressure overload may contribute to the elevation of BNP. Our data also showed that the plasma BNP levels were higher in the moderate–severe PR patients with cardiac symptoms than in those without any symptoms. At our institution, we had performed PVR in symptomatic patients. Therefore, the plasma BNP levels before PVR were higher than those observed in the patients without PVR. Since the plasma BNP level is associated with cardiac symptoms and the severity of the PR and RV functions, it may become a useful predictor for considering the appropriate timing for PVR. Indeed, our data showed that patients with plasma BNP levels higher than 32.15 pg/ml would be more likely to have cardiac symptoms. The plasma BNP level also increases with RV pressure overload due to idiopathic pulmonary arterial hypertension, chronic thromboembolic pulmonary hypertension, and chronic lung disease [12]. Therefore, it is also important to evaluate plasma BNP levels as a continuous follow-up biomarker to predict RV volume expansion or pressure overload. For patients who have elevated plasma BNP levels, we should consider other examinations, such as echocardiography, cardiac catheterization, and CMR, in order to properly assess the timing of PVR. In conclusion, the plasma BNP level is an easy, reliable, and costless parameter for detecting severe PR and RV dysfunction. Measuring the plasma BNP level in patients in the long-term period after repair of TOF may become a useful diagnostic tool to consider the appropriate timing for PVR. Study Limitations Several limitations of this study should be mentioned. First, this is a retrospective study with small number of patients. In particular, only 6 patients had trivial–mild PR and 7 patients underwent PVR. Second, we evaluated the severity of PR using color Doppler echocardiography. However, Li et al. reported that severity of PR and its effects on RV dimensions can be assessed by Doppler echocardiography [13]. Third, we assessed the RV volumes and functions using cardiac catheterization. In recent years, CMR has become the gold standard for measuring the RV function. At our institution, CMR was not performed in CHD patients at the time of this study. We plan to evaluate these patients using CMR in the near future.
Conflict of interest
None.
Pediatr Cardiol
References 1. Abd El Rahman MY, Abdul-Khaliq H, Vogel M, Alexi-Meskishvili V, Gutberlet M, Lange PE (2000) Relation between right ventricular enlargement. QRS duration, and right ventricular function in patients with tetralogy of Fallot and pulmonary regurgitation after surgical repair. Heart 84:416–420 2. Bricker JT (1995) Risks, markers, and causes. Sudden death and tetralogy of Fallot. Circulation 92:158–159 3. Doughan AR, McConnell ME, Lyle TA, Book WM (2005) Effects of pulmonary valve replacement on QRS duration and right ventricular cavity size late after repair of right ventricular outflow tract obstruction. Am J Cardiol 95:1511–1514 4. Eindhoven JA, van den Bosch AE, Jansen PR, Boersma E, RoosHesselink JW (2012) The usefulness of brain natriuretic peptide in complex congenital heart disease: a systematic review. J Am Coll Cardiol 60:2140–2149 5. Frigiola A, Tsang V, Bull C, Coats L, Khambadkone S, Derrick G, Mist B, Walker F, van Doorn C, Bonhoeffer P, Taylor AM (2008) Biventricular response after pulmonary valve replacement for right ventricular outflow tract dysfunction: is age a predictor of outcome? Circulation 118:S182–S190 6. Gatzoulis MA, Till JA, Somerville J, Redinton AN (1995) Mechanoelectrical interaction in tetralogy of Fallot. QRS prolongation relates to right ventricular size and predicts malignant ventricular arrhythmias and sudden death. Circulation 92:231–237 7. Gatzoulis MA, Balaji S, Webber SA, Sui SC, Hokanson JS, Poile C, Rosenthal M, Nakazawa M, Moller JH, Gillette PC, Wevv GD, Redington AN (2000) Risk factors for arrhythmia and sudden cardiac death late after repair of tetralogy of Fallot: a multicentre study. Lancet 356:975–981 8. Gengsakul A, Harris L, Bradley TJ, Webb GD, Williams WG, Siu SC, Merchant N, McCrindle BW (2007) The impact of pulmonary valve replacement after tetralogy of Fallot repair: a matched comparison. Eur J Cardiothorac Surg 32:462–468 9. Geva T, Gauvreau K, Powell AJ, Cecchin F, Rhodes J, Geva J, del Nido P (2010) Randomized trial of pulmonary valve replacement with and without right ventricular remodeling surgery. Circulation 122:S201–S208 10. Henkens IR, van Straten A, Schalij MJ, Hazekamp MG, de Roos A, van der Wall EE, Vliegen HW (2007) Predicting outcome of pulmonary valve replacement in adult tetralogy of Fallot patients. Ann Thorac Surg 83:907–911 11. Khairy P, Landzberg MJ, Gatzoulis MA (2004) Value of programmed ventricular stimulation after tetralogy of fallot repair: a multicenter study. Circulation 109:1994–2000 12. Kucher N, Goldhaber SZ (2003) Cardiac biomarkers for risk stratification of patients with acute pulmonary embolism. Circulation 108:2191–2194 13. Li W, Davlouros PA, Kilner PJ, Pennell DJ, Gibson D, Henein MY, Gatzoulis MA (2004) Doppler-echocardiographic assessment of pulmonary regurgitation in adults with repaired tetralogy of fallot: comparison with cardiovascular magnetic resonance imaging. Am Heart J 147:165–172 14. Lillehei CW, Choen M, Warden HE, Warden HE, Read RC, Aust JB, Dewall RA, Varco RL (1955) Direct vision intracardiac
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
surgical correction of the tetralogy of Fallot, pentalogy of Fallot and pulmonary atresia defects: report of first ten cases. Ann Surg 142:418–442 Mir TS, Falkenberg J, Friedrich B, Gottschalk U, Le TP, Laer S, Weil J (2005) Levels of brain natriuretic peptide in children with right ventricular overload due to congenital cardiac disease. Cardiol Young 15:396–401 Murphy JG, Gersh BJ, Mair DD, Fuster V, McGoon MD, Ilstrup DM, McGoon DC, Kirklin JW, Danielson GK (1993) Long-term outcome in patients undergoing surgical repair of tetralogy of Fallot. N Engl J Med 329:593–599 Norozi K, Buchhorn R, Kaiser C, Hess G, Grunewald RW, Binder L, Wessel A (2005) Plasma N-terminal pro-brain natriuretic peptide as a marker of right ventricular dysfunction in patients with tetralogy of Fallot after surgical repair. Chest 128:2563–2570 Ooi A, Moorjani N, Baliulis G, Keeton BR, Salmon AP, Monro JL, Haw MP (2006) Medium term outcome for infant repair in tetralogy of Fallot: indicators for timing of surgery. Eur J Cardiothorac Surg 30:917–922 Oosterhof T, van Straten A, Vliegen HW, Meijboom FJ, van Dijk AP, Spijkerboer AM, Bouma BJ, Zwinderman AH, Hazekamp MG, de Roos A, Mulder BJ (2007) Preoperative thresholds for pulmonary valve replacement in patients with corrected tetralogy of Fallot using cardiovascular magnetic resonance. Circulation 116:545–551 Oosterhof T, van Straten A, Vliegen HW, Meijboom FJ, van Dijk AP, Spijkerboer AM, Bouma BJ, Zwinderman AH, Hazekamp MG, de Roos A, Mulder BJ (2007) Preoperative thresholds for pulmonary valve replacement in patients with corrected tetralogy of Fallot using cardiovascular magnetic resonance. Circulation 116:545–551 Quattlebaum TG, Varghese PJ, Neill CA, Donahoo JS (1976) Sudden death among postoperative patients with tetralogy of Fallot: a follow-up study of 243 patients for an average of twelve years. Circulation 54:289–293 Saul JP, Alexander ME (1999) Preventing sudden death after repair of tetralogy of Fallot: complex therapy for complex patients. J Cardiovasc Electrophysiol 10:1271–1287 Tang WH, Francis GS, Morrow DA, Newbu LK, Cannon CP, Jesse RL, Storrow AB, Christenson RH, Apple FS, Ravkilde J, Wu AH (2007) National Academy of Clinical Biochemistry Laboratory Medicine. National academy of clinical biochemistry laboratory medicine practice guidelines: clinical utilization of cardiac biomarker testing in heart failure. Circulation 116:e99– e109 Vliegen HW, van Straten A, de Roos A, Roest AA, Schoof PH, Zwinderman AH, Ottenkamp J, van der Wall EE, Hazekamp MG (2002) Magnetic resonance imaging to assess the hemodynamic effects of pulmonary valve replacement in adults late after repair of tetralogy of fallot. Circulation 106:1703–1707 Yeo KT, Wu AH, Apple FS, Kroll MH, Christenson RH, Lewandrowski KB, Sedor FA, Butch AW (2003) Multicenter evaluation of the Roche NT-proBNP assay and comparison to the Biosite Triage BNP assay. Clin Chim Acta 338:107–115
123
ORIGINAL ARTICLE
Clinical Utility of the Plasma Brain Natriuretic Peptide Level in Monitoring Tetralogy of Fallot Patients over the Long Term After Initial Intracardiac Repair: Considerations for Pulmonary Valve Replacement Atsushi Kitagawa • Norihiko Oka • Sumito Kimura • Hisashi Ando • Takashi Honda • Manabu Takanashi • Eri Mineo • Kagami Miyaji • Masahiro Ishii Received: 16 August 2014 / Accepted: 1 December 2014 Ó Springer Science+Business Media New York 2014
Abstract Clinicians are currently encountering an increasing number of patients in the long-term period after tetralogy of Fallot (TOF) repair presenting with pulmonary valve regurgitation (PR) or right ventricular (RV) dysfunction. The purpose of this study was to evaluate the clinical utility of the plasma brain natriuretic peptide (BNP) level and consider surgical indications and timing of pulmonary valve replacement (PVR). We examined 33 patients (21 males, 12 females, mean age 14.5 ± 2.8 years) who underwent TOF repair at Kitasato University Hospital. All patients were evaluated using echocardiography and blood sampling. The mean age at the time of initial repair was 1.3 ± 0.7 years. The patients with moderate–severe PR exhibited significantly higher plasma BNP levels than the patients with trivial–mild PR (mean 37.5 ± 33.1 vs. 17.3 ± 6.6 pg/ml, p = 0.013). The mean plasma BNP level with cardiac symptoms was higher than that observed in the patients without any symptoms (71.4 ± 46.1 vs. 25.0 ± 14.0 pg/ml, p = 0.005). The mean BNP level was significantly decreased after PVR (71.3 ± 46.1–26.1 ± 13.2 pg/ml, p = 0.009), and the plasma BNP level was found to be positively correlated with the RV end-diastolic pressure (r = 0.851; p = 0.008). The optimal BNP cut-off value for considering PVR was 32.15 pg/ml (sensitivity, 85.7 %; specificity, 83.3 %). The plasma BNP level may become a
A. Kitagawa S. Kimura H. Ando T. Honda M. Takanashi E. Mineo M. Ishii Department of Pediatrics, Kitasato University School of Medicine, Sagamihara, Japan N. Oka (&) K. Miyaji Department of Cardiovascular Surgery, Kitasato University School of Medicine, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0374, Japan e-mail: [email protected]
useful diagnostic tool for considering the indications and optimal timing of PVR over the long term after TOF repair. Keywords Tetralogy of Fallot Pulmonary valve regurgitation Pulmonary valve replacement Brain natriuretic peptide Abbreviations BNP Brain natriuretic peptide CHD Congenital heart disease CMR Cardiovascular magnetic resonance imaging EDP End-diastolic pressure EDVI End-diastolic volume index EF Ejection fraction ESVI End-systolic volume index PR Pulmonary valve regurgitation PVR Pulmonary valve replacement RVOTR Right ventricular outflow tract reconstruction TOF Tetralogy of Fallot
Introduction Since the first intracardiac surgery for tetralogy of Fallot (TOF) was reported by Lillehei et al. in 1954 [14], the outcomes of corrective surgery have improved. Clinicians are now encountering an increasing number of patients suffering from late surgical complications, such as heart failure, arrhythmia, and sudden death [21]. For example, Murphy et al. reported that the long-term survival rate of patients treated with TOF repair is lower than that observed in the general population [16]. Pulmonary valve regurgitation (PR) is the primary cause of abnormal hemodynamics over the long term after repair of
123
Pediatr Cardiol
TOF. Severe PR contributes to the development of chronic right ventricular (RV) volume overload and can lead to RV dilatation, RV dysfunction, life-threatening ventricular arrhythmia, and sudden death [6, 7]. There is strong evidence that pulmonary valve replacement (PVR) is highly effective in eliminating or greatly reducing PR [5, 19]. In addition, there is a general consensus that patients exhibiting symptoms related to severe PR and RV dysfunction are candidates for PVR and that physicians should treat these patients before RV dysfunction becomes irreversible. In fact, waiting for symptoms to develop may make it too late to provide adequate treatment. Several studies recommend determining the indications for PVR based on assessments using echocardiography, cardiac catheterization, and cardiovascular magnetic resonance imaging (CMR). However, there are various technical, economic, and invasive problems associated with these diagnostic tools. In recent decades, the level of plasma brain natriuretic peptide (BNP) has received significant attention in the field of heart failure as a marker in the general population [25]. In addition, it is known that the plasma BNP levels are elevated in asymptomatic or only mildly symptomatic patients with chronic RV pressure overload or volume expansion [15, 17], and Eindhoven et al. reported the usefulness of BNP in monitoring patients with complex congenital heart diseases (CHDs), such as TOF [4]. However, the clinical utility of measuring the BNP level to determine the indication for PVR after TOF repair remains controversial. The purpose of this study was, therefore, to evaluate the plasma BNP levels over the long term in patients with a history of TOF repair and identify the indications and optimal surgical timing for PVR.
Patients and Methods Patients This single-institutional study was undertaken at Kitasato University Hospital. Patients older than 10 years of age who had been treated with intracardiac TOF repair were studied.
sampling, including measurement of the plasma BNP level, once or twice a year. In this study, blood was collected in EDTA and lithium heparin vacutainer tubes and immediately sent to the laboratory for an analysis, where the BNP level was assayed. The plasma BNP level was determined using a commercially available chemiluminescence enzyme immunoassay (CLEIA) (MI02 Shionogi BNP kit; Shionogi, Osaka, Japan). The upper normal limit of BNP was 18.4 pg/ml. The severity of PR was assessed based on the width of the vena contracta measured using color Doppler echocardiography. From the color-flow Doppler across the pulmonary valve, the regurgitant jet width was measured at a valve level during early diastole. The pulmonary valve diameter was also measured. The ratio of regurgitant jet width to pulmonary valve diameter was then calculated. We defined moderate–severe PR as PR width/ pulmonary valve ratio C0.5. And we also defined trivial– mild PR as PR width/pulmonary valve ratio \0.5. Cardiac catheterization studies were performed in all seven patients prior to PVR and at 1 year after PVR. We measured both ventricular end-systolic and end-diastolic volumes, and ejection fraction using cardiac catheterization. Statistical Analysis The data are expressed as the median (range) or mean ± standard deviation (SD), as appropriate. Associations between two continuous variables were assessed using a linear regression analysis. Comparisons between paired groups were made using the paired t test or Wilcoxon signed-rank test, as appropriate, while comparisons between independent groups were made using the Mann– Whitney test. A receiver operating characteristic (ROC) curve analysis was performed to determine the BNP cut-off value for predicting symptoms associated with RV dysfunction. The SPSS version 15.0 (SPSS Inc., Chicago, Illinois) software program was used for the statistical analysis, and a p value of \0.05 was considered to be statistically significant.
Results Indication for PVR Clinical Characteristics of the Patients The indications for PVR at our institution included any signs or symptoms attributable to RV dilatation and cardiac dysfunction, such as general fatigue, dyspnea, palpitation and exercise intolerance. Cardiac Studies At our institution, all TOF patients undergo electrocardiography, chest X-ray, echocardiography, and blood
123
Total 33 patients were enrolled. The mean age at the initial intracardiac repair procedure was 1.3 ± 0.7 years, and the mean interval between repair and the examination was 12.8 ± 2.6 years (Table 1). Twenty-seven patients were found to have moderate–severe PR. Six patients were found to have trivial–mild PR. Seven patients underwent PVR due to symptoms associated with cardiac dysfunction (Fig. 1). There were no significant differences between the
Pediatr Cardiol Table 1 Patient characteristics Variable
Table 2 Patient characteristics and severity of PR N = 33
Age at examination (years)
14.5 ± 2.8
Sex (male:female)
21:12
Age at initial repair (years)
1.3 ± 0.7
Interval between initial repair and examination (years)
12.8 ± 2.6
The values are presented as the mean ± standard deviation (SD)
Variable
Moderate– severe PR (n = 27)
Trivial– mild PR (n = 6)
p value
Age at initial repair (years)
1.3 ± 0.7
1.4 ± 0.8
0.310
Age at examination (years)
12.4 ± 5.9
12.9 ± 1.0
0.247
Interval between repair and examination (years)
10.6 ± 5.5
11.5 ± 1.1
0.441
Transannular patch
20
4
Pulmonary valvotomy
4
2
Monocusp valve
3
–
Trivial–mild
21
6
Moderate–severe
6
–
I
23
6
II
3
–
III
1
–
Type of RVOTR
TR grade
NYHA functional class
PR pulmonary valve regurgitation, RVOTR right ventricular outflow tract reconstruction, TR tricuspid valve regurgitation, NYHA New York Heart Association
Fig. 1 Therapeutic strategies employed at our institution PR pulmonary valve regurgitation, PVR pulmonary valve replacement
patients with moderate–severe PR and those with trivial– mild in terms of the mean age at initial repair, mean age at the examination, or interval from repair to the examination. The types of RV outflow tract reconstruction (RVOTR) used in the initial operation are shown in Table 2. Twentynine patients were classified as having a New York Heart Association (NYHA) functional class I status, and four patients were classified as having NYHA functional class II or III status. Relationship Between the Plasma BNP Level and the Severity of PR The mean plasma BNP level in the patients with moderate– severe PR was significantly higher than that observed in the patients with trivial–mild PR (37.5 ± 33.1 vs. 17.3 ± 6.6 pg/ml; p = 0.013) (Fig. 2a). The mean plasma BNP level of patients with symptoms associated with cardiac dysfunction was significantly higher than that observed in the patients having moderate–severe PR without any symptoms (71.4 ± 46.1 vs. 25.0 ± 14.0 pg/ml; p = 0.005) (Fig. 2b).
Fig. 2 Associations between the plasma BNP level and the PR severity and symptoms. a Association between the plasma BNP level and the severity of PR. b Comparison of the plasma BNP levels of moderate–severe PR patients with cardiac symptoms and without
Relationships Between the Plasma BNP Levels and Cardiac Catheterization Parameters Obtained Before PVR Cardiac catheterization was performed in all seven patients prior to PVR. The plasma BNP level before PVR was found to be positively correlated with the RV end-diastolic pressure (EDP) (r = 0.851; p = 0.008), LV EDP (r = 0.755; p = 0.025), and LV end-diastolic volume
123
Pediatr Cardiol
index (EDVI) (r = 0.672; p = 0.049) (Fig. 3b, c, d). Meanwhile, the BNP level was found to be negatively correlated with the RV EDVI (r = -0.680; p = 0.046) (Fig. 3a). In contrast, the RV end-systolic volume index (ESVI), RV ejection fraction (EF), and left ventricular (LV) EF were not found to be correlated with the plasma BNP level (Fig. 3e, f, g, and h). The BNP Cut-Off Value for Considering PVR Figure 4 shows the ROC curve for the plasma BNP level in predicting symptoms associated with RV dysfunction. The area under the curve (AUC) for the plasma BNP level was 0.893, and the optimal cut-off value for the plasma BNP Fig. 3 Relationships between the plasma BNP level the and cardiac catheterization parameters
123
level for considering PVR was 32.15 pg/ml (sensitivity, 85.7 %; specificity, 83.3 %).
Changes One Year After PVR The mean plasma BNP level 1 year after PVR was significantly lower than that observed before PVR (26.1 ± 13.2 vs. 71.4 ± 46.1 pg/ml; p = 0.009) (Fig. 5a). In addition, the QRS duration after PVR was significantly narrower than that measured before PVR (mean 152 ± 12 vs. 167 ± 16 ms; p = 0.009) (Fig. 5b). Among the cardiac catheterization parameters, the mean RV EDVI 1 year after PVR was significantly smaller than that observed before
Pediatr Cardiol
PVR (106.9 ± 14.7 vs. 126.0 ± 19.5 ml/m2; p = 0.008) (Table 3). In all seven patients, the clinical symptoms significantly improved after PVR (Fig. 5c).
Discussion The early and mid-term outcomes of TOF repair have improved in recent years [18]. As the number of long-term survivors has increased, a growing number of late surgical complications have been reported. Numerous studies have shown that patients with a history of TOF repair exhibit many risk factors for exercise intolerance, heart failure, arrhythmia, and sudden death [2, 7, 11, 16, 21, 22]. PR is the primary cause of abnormal hemodynamics after TOF repair, particularly in patients requiring a transannular patch during the procedure. In the present study, our data demonstrated that 27 patients (81.8 %) had moderate– severe PR at a mean of 12.8 ± 2.6 years after the initial Table 3 Changes in the cardiac catheterization parameters before and after PVR Variable
Before PVR
After PVR
p value
2
126.0 ± 19.5
106.9 ± 14.7
0.008
2
RV ESVI (ml/m )
58.2 ± 8.2
52.9 ± 11.6
0.088
RV EDP (mmHg)
6.7 ± 3.3
8.3 ± 4.7
0.058 0.129
RV EDVI (ml/m )
53.1 ± 8.2
50.5 ± 9.0
LV EDVI (ml/m2)
RV EF (%)
105.4 ± 19.6
109.4 ± 21.1
0.199
LV ESVI (ml/m2)
45.3 ± 12.3
45.0 ± 13.8
0.161
LV EDP (mmHg) LV EF (%) Fig. 4 Receiver operating characteristic (ROC) curve analysis of the plasma BNP levels for predicting symptoms associated with RV dysfunction
7.9 ± 1.6
11.2 ± 3.0
0.015
57.0 ± 9.1
59.7 ± 6.0
0.252
RV right ventricle, EDVI end-diastolic volume index, ESVI end-systolic volume index, EDP end-diastolic pressure, EF ejection fraction, LV left ventricle
Fig. 5 Changes in the parameters before and after PVR. a Plasma BNP level. b QRS duration. c NYHA functional class
123
Pediatr Cardiol
intracardiac repair procedure. Severe PR contributes to the development of RV pressure overload and RV dilatation, leading to RV dysfunction. Furthermore, RV dilatation exhibits a close association with prolongation of the QRS duration [1, 3]. PVR is a significantly effective procedure for eliminating or reducing PR [5, 19]. Clinically, several studies have reported significant improvements in the NYHA functional class after PVR [5, 8–10, 20, 24]. In our series, seven patients were treated with PVR. In all of these seven patients, the PR disappeared in association with improvements in clinical symptoms. It is well established that patients with symptoms related to severe PR and RV dysfunction are candidates for PVR. At our institution, we had performed PVR in symptomatic patients, such as general fatigue, palpitation, and exercise intolerance. However, the presence of RV volume overload or dysfunction due to severe PR does not necessarily induce clinical symptoms. Severe PR can be tolerated without any symptoms for many years, while the underlying RV dysfunction becomes irreversible. In fact, waiting for symptoms to develop may make it too late to determine the timing of PVR. Therefore, PVR should be considered before RV dysfunction becomes irreversible. Several studies have suggested the optimal surgical indications for PVR. For example, CMR is a popular modality for measuring the ventricular volume, cardiac function, and valve regurgitation fractions after TOF repair. On the other hand, CMR takes much cost than other examinations. Furthermore, CMR requires a lot of time to measure cardiac parameters. Numerous investigators have attempted to measure the RV function or severity of PR using echocardiography. However, in the absence of significant echocardiographic markers, assessing the RV volume and function after TOF repair remains challenging. The development of other easy, reliable, costless diagnostic tool for detecting cardiac dysfunction in asymptomatic patients with chronic PR is desirable. The plasma BNP level has received significant interest in recent decades. BNP is synthesized and released into the circulation by ventricular myocytes in response to pressure overload, volume expansion, and increased myocardial wall stress. This biomarker exhibits clinical utility for assessing the degree of cardiac impairment. BNP is known to be a well-established marker of heart failure in the general population [23]. The plasma BNP levels are also increased in patients with complex CHD (including TOF, systemic RV, and univentricular hearts) compared with the reference values and/or levels observed in controls, even among asymptomatic patients [4]. The present data showed the plasma BNP levels to be significantly higher in the patients with moderate–severe PR than in those with trivial–mild PR. Among cardiac catheterization parameters, higher RV EDP, LV EDP, and
123
LV EDVI values were found to be associated with a higher BNP level. Severe PR contributes to the development of symptoms associated with RV pressure overload, and biventricular pressure overload may contribute to the elevation of BNP. Our data also showed that the plasma BNP levels were higher in the moderate–severe PR patients with cardiac symptoms than in those without any symptoms. At our institution, we had performed PVR in symptomatic patients. Therefore, the plasma BNP levels before PVR were higher than those observed in the patients without PVR. Since the plasma BNP level is associated with cardiac symptoms and the severity of the PR and RV functions, it may become a useful predictor for considering the appropriate timing for PVR. Indeed, our data showed that patients with plasma BNP levels higher than 32.15 pg/ml would be more likely to have cardiac symptoms. The plasma BNP level also increases with RV pressure overload due to idiopathic pulmonary arterial hypertension, chronic thromboembolic pulmonary hypertension, and chronic lung disease [12]. Therefore, it is also important to evaluate plasma BNP levels as a continuous follow-up biomarker to predict RV volume expansion or pressure overload. For patients who have elevated plasma BNP levels, we should consider other examinations, such as echocardiography, cardiac catheterization, and CMR, in order to properly assess the timing of PVR. In conclusion, the plasma BNP level is an easy, reliable, and costless parameter for detecting severe PR and RV dysfunction. Measuring the plasma BNP level in patients in the long-term period after repair of TOF may become a useful diagnostic tool to consider the appropriate timing for PVR. Study Limitations Several limitations of this study should be mentioned. First, this is a retrospective study with small number of patients. In particular, only 6 patients had trivial–mild PR and 7 patients underwent PVR. Second, we evaluated the severity of PR using color Doppler echocardiography. However, Li et al. reported that severity of PR and its effects on RV dimensions can be assessed by Doppler echocardiography [13]. Third, we assessed the RV volumes and functions using cardiac catheterization. In recent years, CMR has become the gold standard for measuring the RV function. At our institution, CMR was not performed in CHD patients at the time of this study. We plan to evaluate these patients using CMR in the near future.
Conflict of interest
None.
Pediatr Cardiol
References 1. Abd El Rahman MY, Abdul-Khaliq H, Vogel M, Alexi-Meskishvili V, Gutberlet M, Lange PE (2000) Relation between right ventricular enlargement. QRS duration, and right ventricular function in patients with tetralogy of Fallot and pulmonary regurgitation after surgical repair. Heart 84:416–420 2. Bricker JT (1995) Risks, markers, and causes. Sudden death and tetralogy of Fallot. Circulation 92:158–159 3. Doughan AR, McConnell ME, Lyle TA, Book WM (2005) Effects of pulmonary valve replacement on QRS duration and right ventricular cavity size late after repair of right ventricular outflow tract obstruction. Am J Cardiol 95:1511–1514 4. Eindhoven JA, van den Bosch AE, Jansen PR, Boersma E, RoosHesselink JW (2012) The usefulness of brain natriuretic peptide in complex congenital heart disease: a systematic review. J Am Coll Cardiol 60:2140–2149 5. Frigiola A, Tsang V, Bull C, Coats L, Khambadkone S, Derrick G, Mist B, Walker F, van Doorn C, Bonhoeffer P, Taylor AM (2008) Biventricular response after pulmonary valve replacement for right ventricular outflow tract dysfunction: is age a predictor of outcome? Circulation 118:S182–S190 6. Gatzoulis MA, Till JA, Somerville J, Redinton AN (1995) Mechanoelectrical interaction in tetralogy of Fallot. QRS prolongation relates to right ventricular size and predicts malignant ventricular arrhythmias and sudden death. Circulation 92:231–237 7. Gatzoulis MA, Balaji S, Webber SA, Sui SC, Hokanson JS, Poile C, Rosenthal M, Nakazawa M, Moller JH, Gillette PC, Wevv GD, Redington AN (2000) Risk factors for arrhythmia and sudden cardiac death late after repair of tetralogy of Fallot: a multicentre study. Lancet 356:975–981 8. Gengsakul A, Harris L, Bradley TJ, Webb GD, Williams WG, Siu SC, Merchant N, McCrindle BW (2007) The impact of pulmonary valve replacement after tetralogy of Fallot repair: a matched comparison. Eur J Cardiothorac Surg 32:462–468 9. Geva T, Gauvreau K, Powell AJ, Cecchin F, Rhodes J, Geva J, del Nido P (2010) Randomized trial of pulmonary valve replacement with and without right ventricular remodeling surgery. Circulation 122:S201–S208 10. Henkens IR, van Straten A, Schalij MJ, Hazekamp MG, de Roos A, van der Wall EE, Vliegen HW (2007) Predicting outcome of pulmonary valve replacement in adult tetralogy of Fallot patients. Ann Thorac Surg 83:907–911 11. Khairy P, Landzberg MJ, Gatzoulis MA (2004) Value of programmed ventricular stimulation after tetralogy of fallot repair: a multicenter study. Circulation 109:1994–2000 12. Kucher N, Goldhaber SZ (2003) Cardiac biomarkers for risk stratification of patients with acute pulmonary embolism. Circulation 108:2191–2194 13. Li W, Davlouros PA, Kilner PJ, Pennell DJ, Gibson D, Henein MY, Gatzoulis MA (2004) Doppler-echocardiographic assessment of pulmonary regurgitation in adults with repaired tetralogy of fallot: comparison with cardiovascular magnetic resonance imaging. Am Heart J 147:165–172 14. Lillehei CW, Choen M, Warden HE, Warden HE, Read RC, Aust JB, Dewall RA, Varco RL (1955) Direct vision intracardiac
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
surgical correction of the tetralogy of Fallot, pentalogy of Fallot and pulmonary atresia defects: report of first ten cases. Ann Surg 142:418–442 Mir TS, Falkenberg J, Friedrich B, Gottschalk U, Le TP, Laer S, Weil J (2005) Levels of brain natriuretic peptide in children with right ventricular overload due to congenital cardiac disease. Cardiol Young 15:396–401 Murphy JG, Gersh BJ, Mair DD, Fuster V, McGoon MD, Ilstrup DM, McGoon DC, Kirklin JW, Danielson GK (1993) Long-term outcome in patients undergoing surgical repair of tetralogy of Fallot. N Engl J Med 329:593–599 Norozi K, Buchhorn R, Kaiser C, Hess G, Grunewald RW, Binder L, Wessel A (2005) Plasma N-terminal pro-brain natriuretic peptide as a marker of right ventricular dysfunction in patients with tetralogy of Fallot after surgical repair. Chest 128:2563–2570 Ooi A, Moorjani N, Baliulis G, Keeton BR, Salmon AP, Monro JL, Haw MP (2006) Medium term outcome for infant repair in tetralogy of Fallot: indicators for timing of surgery. Eur J Cardiothorac Surg 30:917–922 Oosterhof T, van Straten A, Vliegen HW, Meijboom FJ, van Dijk AP, Spijkerboer AM, Bouma BJ, Zwinderman AH, Hazekamp MG, de Roos A, Mulder BJ (2007) Preoperative thresholds for pulmonary valve replacement in patients with corrected tetralogy of Fallot using cardiovascular magnetic resonance. Circulation 116:545–551 Oosterhof T, van Straten A, Vliegen HW, Meijboom FJ, van Dijk AP, Spijkerboer AM, Bouma BJ, Zwinderman AH, Hazekamp MG, de Roos A, Mulder BJ (2007) Preoperative thresholds for pulmonary valve replacement in patients with corrected tetralogy of Fallot using cardiovascular magnetic resonance. Circulation 116:545–551 Quattlebaum TG, Varghese PJ, Neill CA, Donahoo JS (1976) Sudden death among postoperative patients with tetralogy of Fallot: a follow-up study of 243 patients for an average of twelve years. Circulation 54:289–293 Saul JP, Alexander ME (1999) Preventing sudden death after repair of tetralogy of Fallot: complex therapy for complex patients. J Cardiovasc Electrophysiol 10:1271–1287 Tang WH, Francis GS, Morrow DA, Newbu LK, Cannon CP, Jesse RL, Storrow AB, Christenson RH, Apple FS, Ravkilde J, Wu AH (2007) National Academy of Clinical Biochemistry Laboratory Medicine. National academy of clinical biochemistry laboratory medicine practice guidelines: clinical utilization of cardiac biomarker testing in heart failure. Circulation 116:e99– e109 Vliegen HW, van Straten A, de Roos A, Roest AA, Schoof PH, Zwinderman AH, Ottenkamp J, van der Wall EE, Hazekamp MG (2002) Magnetic resonance imaging to assess the hemodynamic effects of pulmonary valve replacement in adults late after repair of tetralogy of fallot. Circulation 106:1703–1707 Yeo KT, Wu AH, Apple FS, Kroll MH, Christenson RH, Lewandrowski KB, Sedor FA, Butch AW (2003) Multicenter evaluation of the Roche NT-proBNP assay and comparison to the Biosite Triage BNP assay. Clin Chim Acta 338:107–115
123
