B-type natriuretic peptide to assess haemodynamic status after cardiac surgery

Transcription

1 British Journal of Anaesthesia 97 (6): (2006) doi: /bja/ael271 Advance Access publication October 17, 2006 B-type natriuretic peptide to assess haemodynamic status after cardiac surgery A. Mekontso-Dessap 1 *, L. Tual 2, M. Kirsch 2, G. D Honneur 2, D. Loisance 2, L. Brochard 1 and J.-L. Teboul 3 1 Medical Intensive Care Unit and 2 Cardiac Surgery Intensive Care Unit, Université Paris XII, Centre Hospitalier Universitaire Henri Mondor, Assistance Publique Hôpitaux de Paris, INSERM U 651, 51 avenue du Maréchal de Lattre de Tassigny, Créteil Cedex, France. 3 Medical Intensive Care Unit, Université Paris XI, Centre Hospitalier Universitaire de Bicêtre, Assistance Publique Hôpitaux de Paris, 78, rue du Général Leclerc, Le Kremlin-Bicêtre Cedex, France *Corresponding author: Service de Réanimation Médicale, Centre Hospitalier Universitaire Henri Mondor, 51, avenue du M al de Lattre de Tassigny; Créteil Cedex, France. armand.dessap@creteil.inserm.fr Background. B-type natriuretic peptide (BNP) is the most powerful hormonal marker of left ventricular dysfunction and could be considered as an indicator of ventricular preload. The aim of this prospective study was to compare the respective value of BNP and cardiac filling pressures to assess the response to volume load after cardiac surgery. Methods. Thirty-seven mechanically ventilated patients suffering from acute circulatory failure immediately after cardiac surgery, and equipped with a pulmonary-artery catheter were included. All haemodynamic measurements were taken before and after volume expansion using 500 ml of 4% modified fluid gelatin. Results. Fifteen patients were volume responders (CI increase>15%) and 22 were nonresponders. Right atrial pressure, pulmonary-artery occlusion pressure and BNP before volume loading were not significantly different between the responders and non-responders. BNP concentration before volume infusion significantly correlated to preoperative left ventricular ejection fraction, aortic cross-clamping time, serum creatinine, mean pulmonary arterial pressure and intensive care unit duration whereas no correlation was found with pulmonary-artery occlusion pressure or cardiac index. Conclusion. BNP level after cardiac surgery was influenced by many perioperative variables, limiting its usefulness as an indicator of cardiac preload or a predictor of volume responsiveness in this population. Br J Anaesth 2006; 97: Keywords: complications, hypotension; natriuretic peptide; surgery, cardiac; vascular pressures Accepted for publication: July 5, 2006 Haemodynamic instability is frequent in the immediate course of cardiac surgery, with various causes including hypovolaemia, cardiac dysfunction, and loss of vascular responsiveness. The use of pulmonary-artery catheter (PAC) in the postoperative period is seriously questioned as no benefit of PAC-directed therapy has been evidenced in this setting. 1 In particular, right atrial pressure and pulmonary-artery occlusion pressure were shown to be of little value in predicting fluid responsiveness. 2 B-type natriuretic peptide (BNP) is recognized as the most powerful hormonal marker of left ventricular dysfunction. As this 32-amino acid protein is released from the cardiac ventricles in response to myocyte stretch, 3 it could be considered as a marker of ventricular preload. The aim of this clinical study was to compare the respective value of BNP and cardiac filling pressures to assess volume responsiveness after cardiac surgery. Patients and methods The study was approved by the institutional Ethics Committee of the Société de Réanimation de Langue Française, Ó The Board of Management and Trustees of the British Journal of Anaesthesia All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Mekontso-Dessap et al. that waived the need for written informed consent because blood samples obtained for routine follow-up were used, according to French legislation. However, all patients received written and oral information about the study. We included mechanically ventilated patients immediately after cardiac surgery. Inclusion criteria were acute circulatory failure defined by a systolic blood pressure <90 mm Hg or the need for vasopressors, and placement of a PAC. Patients were excluded in case of severe hypothermia (body temperature<34.0 C) or severe renal failure (creatinine clearance<30 ml min 1 ). Creatinine clearance was estimated using the Cockroft and Gault formula. 4 Type of surgery included Coronary Artery Bypass Grafting (CABG, n=18), valve procedures (n=11), CABG+valve procedures (n=5) and other procedures (n=3). The anaesthetic protocol was identical in all patients and included i.v. midazolam, propofol, pancuronium and fentanyl or sulfentanyl. Cardiopulmonary bypass was conducted under moderate systemic hypothermia (28 32 C), with non-pulsatile flow, centrifugal pump and gravity venous drainage. A hollowfibre membrane oxygenator was used. Antegrade cold crystalloid cardioplegia was used for myocardial protection. Patients were included within less than 3 h after the end of cardiac surgery. At the time of haemodynamic measurement, patients were receiving epinephrine (n=6), dobutamine (n=19), norepinephrine (n=2) or no vasoactive drug (n=10). All patients were mechanically ventilated using an assist controlled mode with a tidal volume of 9.3 (0.9) ml kg 1, a respiratory frequency of 15 (2) breaths min 1 and a positive end-expiratory pressure of 4.8 (1.1) cm H 2 O. Nineteen patients underwent continuous sedation using propofol, and no patient was paralysed. Patients were studied while supine, and the pressure measurements were made with reference to the midaxillary line. Right atrial pressure and pulmonary-artery occlusion pressure were recorded throughout the respiratory cycle and measured at end-expiration. Cardiac output was calculated as the mean of five measurements obtained by injecting 10 ml of dextrose solution randomly during the respiratory cycle. Cardiac index, stroke volume index, vascular resistances and stroke work indices were calculated using standard formulae. All haemodynamic measurements were performed before and after volume expansion using 500 ml of 4% modified fluid gelatin delivered in min. Ventilatory settings and dosages of inotropic and vasopressive drugs were held constant. BNP was measured at the time of haemodynamic measurements, before volume expansion for all patients and after volume expansion for 18 patients. This quantitative determination was performed on blood samples collected into tubes containing EDTA, using a fluorescence immunoassay kit (Triage, Biosite Diagnostics). Statistical analysis was performed using SPSS Base 11.5 statistical software (SPSS Inc., Chicago, IL). Continuous variables were expressed as mean (SD). The Mann Whitney U-test was used to compare unpaired values before volume expansion. Linear correlations were tested using the Pearson method. The paired samples t-test or the Wilcoxon test were used to compare paired values before and after volume expansion. Patients were separated in two groups depending on the change in cardiac index following volume expansion: responders (cardiac index increase>15%) and nonresponders (cardiac index increase<15%). We chose this cut-off value in reference to previous studies that used cardiac index to assess the response to fluid infusion. 256 All these studies chose a benchmark of 15% because this cut-off value was found to be above the minimal difference between two measures of cardiac output by thermodilution to suggest clinical significance. 7 A repeated measures analysis of variance was performed using cardiac index as the outcome variable, and right atrial pressure, pulmonary-artery occlusion pressure and (log transformed) BNP as covariates. In addition, a multivariate regression analysis using the change in cardiac index as the outcome variable and right atrial pressure, pulmonary-artery occlusion pressure and baseline (log transformed) BNP as predictor variables was also performed. A two-tailed P-value of less than 0.05 was used to indicate statistical significance. Results Thirty-seven patients (19 males and 18 females) were included in the study. Their main characteristics and surgery-related variables are listed in Table 1. All patients but two were discharged alive from the hospital. Table 2 shows haemodynamic data and BNP levels before and after volume expansion. It has to be noted that fluid infusion resulted in a significant increase in right atrial pressure, pulmonary-artery occlusion pressure, cardiac index and stroke volume index while BNP concentration did not change. There was no correlation between change in BNP and change in CI (r= 0.08, P=0.74). Fifteen patients were volume responders (cardiac index increase>15%) and 22 were non-responders (cardiac index increase<15%). Right atrial and pulmonary-artery occlusion pressures before volume loading were not significantly different between responders and non-responders [13 (6) vs Table 1 Patient characteristics, surgery-related variables and their correlation with baseline B-type natriuretic peptide (BNP). Values are given as mean (SD) Variables Value Correlation with baseline BNP, r (P-value) Age (yr) 70.2 (9.9) 0.23 (0.16) Weight (kg) 71 (13) 0.06 (0.74) Serum creatinine (mmol litre 1 ) 90 (9) 0.40 (0.01) Creatinine clearance (ml min 1 ) 68 (28) 0.29 (0.08) Preoperative left ventricle ejection fraction (%) 54 (17) 0.43 (0.01) Cardiopulmonary bypass time (min) 138 (66) 0.14 (0.43) Aortic cross-clamping time (min) 83 (38) 0.33 (0.05) Mechanical ventilation duration (days) 1.9 (3.3) 0.23 (0.20) Intensive care unit duration (days) 4.9 (5.2) 0.41 (0.01) 778

3 BNP and haemodynamics after heart surgery Table 2 Effects of volume expansion on haemodynamic variables and correlation of haemodynamic variables with BNP at baseline. Baseline measurements were made <3 h after the end of cardiac surgery; volume expansion was performed using 500 ml of 4% modified fluid gelatin delivered in min; BNP, B-type natriuretic peptide; *P<0.05 as compared with baseline value; **P<0.01 as compared with baseline value. Values at baseline and after volume expansion are given as mean (SD) Variable Value at baseline Value after volume expansion Correlation with BNP at baseline, r (P-value) Heart rate 92 (20) 90 (17) 0.01 (0.97) (beats min 1 ) Mean arterial pressure 84 (12) 82 (15) 0.23 (0.18) Mean pulmonary 27 (10) 28 (7) 0.40 (0.02) arterial pressure Right atrial pressure 15 (6) 16 (6)* 0.17 (0.32) Pulmonary-artery 15 (7) 17 (6)* 0.18 (0.28) occlusion pressure Cardiac index 2.8 (0.8) 3.1 (0.9)** 0.03 (0.86) (litre min 1 m 2 ) Stroke volume 31 (9) 35 (8)** 0.03 (0.87) index (ml m 2 ) Systemic vascular 2163 (790) 1800 (691)** 0.19 (0.25) resistance index (dynes cm 5 m 2 ) Pulmonary vascular 362 (176) 314 (115)** 0.34 (0.04) resistance index (dynes cm 5 m 2 ) Left ventricle stroke 30 (11) 31 (12) 0.11 (0.52) work index (g m 2 ) Right ventricle stroke 5.1 (3.3) 5.7 (3.1) 0.35 (0.04) work index (g m 2 ) BNP (pg ml 1 ) 364 (507) 318 (327) 16 (5) mm Hg, P=0.11 and 14 (10) vs 16 (5) mm Hg, P=0.13 respectively]. The BNP levels before volume expansion were not significantly different between responders and non-responders to fluid infusion [264 (294) vs 389 (531) pg ml 1, P=0.73, Fig. 1]. In the multivariate regression analysis, none of the factors tested (right atrial pressure, pulmonary-artery occlusion pressure and BNP) was significantly associated with the change in cardiac index (P=0.07, P=0.07 and P=0.09 respectively). Similarly, in the repeated measures analysis of variance, none of these variables was significantly associated with cardiac index (P=0.43, P=0.33 and P=0.75 respectively). Table 1 displays correlations between pre-infusion BNP levels and clinical and surgery-related variables. The plasma BNP concentration before volume infusion significantly correlated with preoperative left ventricular ejection fraction (Fig. 2), aortic cross-clamping time, serum creatinine and intensive care unit duration. There was no significant difference in baseline BNP level between men and women [439 (582) vs 232 (214) pg ml 1, P=0.66]. Table 2 displays correlations between plasma BNP levels and haemodynamic variables measured before volume expansion. The preinfusion plasma BNP level significantly correlated with BNP before VE, pg ml Responders Non responders Fig 1 BNP level before volume expansion in responders and nonresponders. Baseline measurements were made at <3 h after the end of cardiac surgery; BNP, B-type natriuretic peptide; VE, volume expansion. BNP before VE, pg ml r = 0.43 P=0.01 Preoperative LVEF, % Fig 2 Correlation between preoperative left ventricle ejection fraction and baseline BNP level. LVEF, left ventricle ejection fraction; BNP, B-type natriuretic peptide; VE, volume expansion. mean pulmonary arterial pressure and pulmonary vascular resistance but not with pulmonary-artery occlusion pressure, right atrial pressure and cardiac index (Fig. 3). Discussion Our results demonstrate the limited usefulness of BNP as well as cardiac filling pressures to assess fluid responsiveness immediately after cardiac surgery. We found that BNP level was not a better predictor of volume responsiveness 779

4 Mekontso-Dessap et al. BNP before VE, pg ml r =0.18 P= Pulmonary artery occlusion pressure, mm Hg Fig 3 Correlation between pulmonary-artery occlusion pressure and baseline BNP level. Baseline measurements were made at <3 h after the end of cardiac surgery; BNP, B-type natriuretic peptide; VE, volume expansion. than cardiac filling pressures as BNP concentration before volume loading was not different in responders and nonresponders. Because BNP is released from the ventricles in response to myocyte stretch, it could be a marker of cardiac preload. However, in our study, BNP did not seem a good marker of cardiac preload, as volume infusion did not increase BNP levels while stroke volume, right atrial pressure and pulmonary-artery occlusion pressure increased with fluid infusion. In addition, BNP levels did not correlate with cardiac filling pressures in our patients. The fact that BNP level did not change after volume infusion is consistent with the study of Matsumoto and colleagues 8 who reported a non-significant increase in plasma BNP concentration despite the raised left ventricular end-diastolic pressure in the early phase of ventricular overload. A possible explanation is the particular secretion profile of BNP. In fact, contrary to preformed ANP which is released rapidly from secretory granules in atrial myocytes, 9 BNP is regulated at the level of gene expression and therefore, its release kinetics are delayed 10 with a biologically half-life time of about 20 min. Patients could therefore have increased concentrations of precursors while lacking biologically active peptides. The lack of correlation between cardiac filling pressures and BNP levels is in discordance with previous studies reporting a close relationship between BNP and cardiac filling pressures in patients with heart failure One reason for this discrepancy may be related to the differences in the studied populations. Indeed, cardiac surgical procedures per se seem to affect BNP secretion as suggested by studies that demonstrated BNP release after aortic declamping. In this regard, we found a significant correlation between BNP level and aortic cross-clamp time confirming the results of Morimoto and colleagues. 15 In another study, BNP release after cardiac reperfusion was reported to be correlated with the severity of ischaemia (as assessed by myocardial lactate production), and to be higher in patients with postoperative complications as compared with uneventful patients. 17 In the same setting, Hutfless and colleagues 18 demonstrated that elevated peak postoperative BNP levels were associated with prolonged hospital stay and mortality within 1 yr. Incidentally, in our study, BNP levels correlated with the intensive care unit duration. Numerous previous studies in critically ill patients as well as in cardiac surgery patients evidenced the poor value of right atrial pressure and pulmonary-artery occlusion pressure to assess volume responsiveness. Since the publication of these studies and of recent review articles on the issue of volume responsiveness, we do not take decisions to perform fluid challenge on the basis of values of cardiac filling pressures any more. Our present study has confirmed that neither baseline right atrial pressure nor pulmonary-artery occlusion pressure were good predictors of the haemodynamic response to volume. One important reason explaining this finding is that ventricular filling pressures are poor markers of ventricular preload. 2 In this regard, these pressures and their changes are known to be less correlated with stroke volume than ventricular end-diastolic volumes and their changes in normal subjects, 27 critically ill and cardiac surgical patients. Furthermore, even a good marker of preload is not automatically a reliable predictor of preload responsiveness 2 as demonstrated in critically ill patients or cardiac surgical patients in whom volumetric markers of preload were measured. 23 This could be explained by the great heterogeneity of Frank Starling function curves among patients. In this regard, for a given baseline preload marker value, the increase in preload after volume expansion would result in a larger increase in stroke volume in the normal heart than in the failing heart. Pulmonary hypertension is of common occurrence after cardiopulmonary bypass. 30 In this regard, mean pulmonaryartery pressure was above 25 mm Hg in our patients. We also found a correlation between BNP levels and mean pulmonary arterial pressure, pulmonary vascular resistance index, and right ventricle stroke work index. Our results agree with those of previous studies in patients with pulmonary hypertension that reported significant correlations between BNP levels and mean pulmonary arterial pressure, as well as with pulmonary vascular resistance. BNP level significantly correlated with serum creatinine concentration. One possible explanation is that BNP removal from the circulation implies inactivation by vascular and renalendopeptidases. Studies examining BNP levels in patients with renal insufficiency have typically reported that higher BNP levels are associated with reduced estimated 780

5 BNP and haemodynamics after heart surgery glomerular filtration rates However, in our study, there was no significant linear correlation between BNP and creatinine clearance. Our study has some limitations. Firstly, we cannot exclude a type II statistical error. The small number of patients and the wide standard deviations of BNP values associated with a non-normal distribution may explain the negative results. However, non-parametric tests were used for analysis. Furthermore, when using a transformation of BNP (LnBNP or logbnp), or a repeated measures analysis of variance, the results were not altered. Secondly, we used volume expansion as therapeutic intervention. Therefore, we cannot exclude that some degree of haemodilution might have influenced the post-infusion BNP levels. In conclusion, plasma BNP level after cardiac surgery is influenced by perioperative variables such as aortic crossclamp time and pulmonary hypertension that precludes its use as a predictor of volume responsiveness in this population. Acknowledgement Support was provided solely from institutional and departmental sources. References 1 Sandham JD, Hull RD, Brant RF, et al. A randomized, controlled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003; 348: Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest 2002; 121: de Lemos JA, McGuire DK, Drazner MH. B-type natriuretic peptide in cardiovascular disease. Lancet 2003; 362: Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 16: De Backer D, Heenen S, Piagnerelli M, Koch M, Vincent JL. Pulse pressure variations to predict fluid responsiveness: influence of tidal volume. Intensive Care Med 2005; 31: Preisman S, Kogan S, Berkenstadt H, Perel A. Predicting fluid responsiveness in patients undergoing cardiac surgery: functional haemodynamic parameters including the Respiratory Systolic Variation Test and static preload indicators. Br J Anaesth 2005; 95: Stetz CW, Miller RG, Kelly GE, Raffin TA. Reliability of the thermodilution method in the determination of cardiac output in clinical practice. Am Rev Respir Dis 1982; 126: Matsumoto N, Akaike M, Nishiuchi T, Kawai H, Saito S. Different secretion profiles of atrial and brain natriuretic peptides after acute volume loading in patients with ischemic heart disease. Acta Cardiol 1997; 52: Hasegawa K, Fujiwara H, Itoh H, et al. Light and electron microscopic localization of brain natriuretic peptide in relation to atrial natriuretic peptide in porcine atrium. Immunohistocytochemical study using specific monoclonal antibodies. Circulation 1991; 84: de Bold AJ, Ma KK, Zhang Y, de Bold ML, Bensimon M, Khoshbaten A. The physiological and pathophysiological modulation of the endocrine function of the heart. Can J Physiol Pharmacol 2001; 79: Friedl W, Mair J, Thomas S, Pichler M, Puschendorf B. Relationship between natriuretic peptides and hemodynamics in patients with heart failure at rest and after ergometric exercise. Clin Chim Acta 1999; 281: Maeda K, Tsutamoto T, Wada A, Hisanaga T, Kinoshita M. Plasma brain natriuretic peptide as a biochemical marker of high left ventricular end-diastolic pressure in patients with symptomatic left ventricular dysfunction. Am Heart J 1998; 135: Cheng V, Kazanagra R, Garcia A, et al. A rapid bedside test for B-type peptide predicts treatment outcomes in patients admitted for decompensated heart failure: a pilot study. J Am Coll Cardiol 2001; 37: Matsumoto A, Hirata Y, Momomura S, et al. Effects of exercise on plasma level of brain natriuretic peptide in congestive heart failure with and without left ventricular dysfunction. Am Heart J 1995; 129: Morimoto K, Mori T, Ishiguro S, Matsuda N, Hara Y, Kuroda H. Perioperative changes in plasma brain natriuretic peptide concentrations in patients undergoing cardiac surgery. Surg Today 1998; 28: Georges A, Forestier F, Valli N, Plogin A, Janvier G, Bordenave L. Changes in type B natriuretic peptide (BNP) concentrations during cardiac valve replacement. Eur J Cardiothorac Surg 2004; 25: Mair P, Mair J, Bleier J, Hormann C, Balogh D, Puschendorf B. Augmented release of brain natriuretic peptide during reperfusion of the human heart after cardioplegic cardiac arrest. Clin Chim Acta 1997; 261: Hutfless R, Kazanegra R, Madani M, et al. Utility of B-type natriuretic peptide in predicting postoperative complications and outcomes in patients undergoing heart surgery. J Am Coll Cardiol 2004; 43: Wagner JG, Leatherman JW. Right ventricular end-diastolic volume as a predictor of the hemodynamic response to a fluid challenge. Chest 1998; 113: Tavernier B, Makhotine O, Lebuffe G, Dupont J, Scherpereel P. Systolic pressure variation as a guide to fluid therapy in patients with sepsis-induced hypotension. Anesthesiology 1998; 89: Michard F, Boussat S, Chemla D, et al. Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med 2000; 162: Tousignant CP, Walsh F, Mazer CD. The use of transesophageal echocardiography for preload assessment in critically ill patients. Anesth Analg 2000; 90: Rex S, Brose S, Metzelder S, et al. Prediction of fluid responsiveness in patients during cardiac surgery. Br J Anaesth 2004; 93: Kramer A, Zygun D, Hawes H, Easton P, Ferland A. Pulse pressure variation predicts fluid responsiveness following coronary artery bypass surgery. Chest 2004; 126: Bendjelid K, Romand JA. Fluid responsiveness in mechanically ventilated patients: a review of indices used in intensive care. Intensive Care Med 2003; 29: Coudray A, Romand JA, Treggiari M, Bendjelid K. Fluid responsiveness in spontaneously breathing patients: a review of indexes used in intensive care. Crit Care Med 2005; 33: Kumar A, Anel R, Bunnell E, et al. Pulmonary artery occlusion pressure and central venous pressure fail to predict ventricular filling volume, cardiac performance, or the response to volume infusion in normal subjects. Crit Care Med 2004; 32: Diebel LN, Wilson RF, Tagett MG, Kline RA. End-diastolic volume. A better indicator of preload in the critically ill. Arch Surg 1992; 127:

6 Mekontso-Dessap et al. 29 GodjeO, PeyerlM, SeebauerT, LammP, MairH, ReichartB. Central venous pressure, pulmonary capillary wedge pressure and intrathoracic blood volumes as preload indicators in cardiac surgery patients. Eur J Cardiothorac Surg 1998; 13: 533 9; discussion Riedel B. The pathophysiology and management of perioperative pulmonary hypertension with specific emphasis on the period following cardiac surgery. Int Anesthesiol Clin 1999; 37: Nagaya N, Nishikimi T, Uematsu M, et al. Plasma brain natriuretic peptide as a prognostic indicator in patients with primary pulmonary hypertension. Circulation 2000; 102: Ishii J, Nomura M, Ito M, et al. Plasma concentration of brain natriuretic peptide as a biochemical marker for the evaluation of right ventricular overload and mortality in chronic respiratory disease. Clin Chim Acta 2000; 301: Sarnak MJ. Cardiovascular complications in chronic kidney disease. Am J Kidney Dis 2003; 41: McCullough PA, Duc P, Omland T, et al. B-type natriuretic peptide and renal function in the diagnosis of heart failure: an analysis from the Breathing Not Properly Multinational Study. Am J Kidney Dis 2003; 41: