Secondary Logo

Journal Logo

B-Type Natriuretic Peptide Levels and Continuous-Flow Left Ventricular Assist Devices

Sareyyupoglu, Basar*; Boilson, Barry A.; Durham, Lucian A.*; McGregor, Christopher G. A.*; Daly, Richard C.*; Redfield, Margaret M.; Edwards, Brooks S.; Frantz, Robert P.; Pereira, Naveen L.; Park, Soon J.*

doi: 10.1097/MAT.0b013e3181f127a7
Clinical Cardiovascular/Cardiopulmonary Bypass
Free

We postulated that postoperative B-type natriuretic peptide (BNP) levels would be reflective of the degree of hemodynamic support rendered by various pump speeds settings (RPM) of continuous-flow left ventricular assist devices (LVADs). Twenty LVAD patients were evaluated prospectively (Jarvik 2000: n = 9, HeartMate II: n = 11). The mean age was 57.7 ± 14.9 years, and 14 were male. B-type natriuretic peptide levels were drawn while the patients were supported on LVADs at variable RPM settings. The RPM settings were correlated with the changes in BNP levels. Eleven patients underwent LVAD implantation for a lifelong support while the rest were as a bridge therapy to transplantation. Four patients required LVAD change out for various causes of pump failure. Postoperative BNP levels decreased dramatically with the initiation of LVAD support. The levels correlated inversely with the degree of hemodynamic support rendered at various RPM settings of the HeartMate II (p < 0.001). Overall, BNP levels decreased significantly in 2 days after RPM increase. We observed a significant inverse correlation between the postoperative BNP levels and the degree of LVAD support. The effective LVAD support seems to result in a marked reduction in BNP levels, and monitoring serial BNP levels may be helpful in managing patients supported on continuous LVAD.

From the Divisions of *Cardiovascular Surgery and †Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota.

Submitted for consideration February 2010; accepted for publication in revised form June 2010.

Reprint Requests: Soon J. Park, MD, Division of Cardiovascular Surgery, Mayo Clinic, Joseph 5-200, Rochester, MN 55205. Email: park.soon@mayo.edu..

Plasma B-type natriuretic peptide (BNP) is transcribed and translated within the ventricular myocardium.1 Plasma BNP levels seem to reflect the degree of ventricular dysfunction in patients with advanced congestive heart failure (CHF), and high levels are associated with a poor survival.2,3 Implantation of a pulsatile left ventricular assist device has proven to be a lifesaving measure for many patients for it can effectively restore circulation and unload the failing left ventricle.4,5 Pulse rates and pressures generated by a volume displacement pump are similar to those generated by a normal heart. Therefore, vital signs of the pulsatile LVAD patients remain as useful clinical parameters to monitor. However, with a newer generation continuous flow pump, the mechanism of blood propulsion is entirely different. The patients supported on continuous LVAD no longer have measurable pulse rates or pressures. Echocardiography can be helpful in assessing the dimensions of the left ventricle while adjusting the pump speed for a proper unloading of the left ventricle.6,7 Nonetheless, we are clearly entering a new era of uncharted physiology with the continuous-flow LVAD. The optimal continuous LVAD support settings are yet to be determined. Meanwhile, it would be helpful to find a biological marker that would reflect the patient's physiological response to the continuous LVAD support. Therefore, we embarked on this study to see whether BNP would be a useful physiological biomarker and to investigate the relationship of BNP with different speed levels.

Back to Top | Article Outline

Methods

Patient Selection and Data Collection

Institutional approval for this study was obtained from Mayo Clinic Institutional Review Board (IRB), and patients gave their informed consent to participate in this clinical research study. Twenty patients were included in the study to evaluate the relationship between various pump speeds and BNP levels after LVAD implantation. Only the patients who underwent implantation of a continuous LVAD between February 2007 and May 2008 were included. Eleven patients were supported on the HeartMate II pumps while nine patients were on Jarvik 2000. Jarvik 2000 pumps were used as bridge therapy to transplantation (BTT) devices, whereas the HeartMate II pumps were used as destination therapy (DT) devices. Plasma samples for standard blood chemistry and BNP were obtained preoperatively and at regular intervals after LVAD implantation (daily until discharge, then monthly, and at readmissions). All patients underwent scheduled echocardiographic evaluations to assess the degree of ventricular unloading on LVAD support.

Back to Top | Article Outline

Quantitative Analysis of BNP

Biochemical analyses for BNP levels were performed on the UniCel® DxI 800 Access® Immunoassay System from Beckman Coulter (Fullerton, CA). The intra-assay variability was 6%–8%. Investigators who were blinded to the sample source performed the biochemical analyses.

B-type natriuretic peptide and creatinine levels results were collected prospectively. Adjustments in LVAD RPM settings were made by the primary VAD service based on the team's assessment of each patient's clinical and echocardiographic findings. An independent investigator who did not participate in the patient care recorded the pump speed settings, the corresponding plasma BNP, and creatinine levels. The pump speed settings used for this study patients ranged from 2 to 5 for Jarvik 2000 and 8,400–10,600 RPM for the HeartMate II. One patient who underwent a device change out from Jarvik 2000 to the HeartMate II during the same hospital stay was evaluated for BNP changes under each device.

Back to Top | Article Outline

Statistical Analysis

Data were entered in an Excel spreadsheet (Microsoft Corporation, Redmond, WA) and transferred to a SAS file (SAS Institute Inc., Cary, NC) for data description and analyses. The continuous variables were expressed as a mean (±SD) or a median (±range) where appropriate, whereas the categorical variables were expressed as a frequency or a percentage. Differences in echocardiographic values over time were evaluated using the repeated measures analyses of variance with Tukey ad hoc tests for comparisons at specific time points. B-type natriuretic peptide and creatinine levels were correlated with RPM settings in exploratory analyses using Spearman (nonparametric) correlation coefficients. Repeated measures of variance were used to evaluate BNP change after RPM alteration. A p value <0.05 was considered statistically significant for all analyses.

Back to Top | Article Outline

Results

Demographic characteristics of the patients are summarized in Table 1. Ischemic (49%) and idiopathic dilated (35%) cardiomyopathy comprised the majority of heart failure etiology. Other causes of heart failure included alcoholic, hypertrophic, postpartum, and chemotherapy-induced cardiomyopathy. One patient in the BTT group died in the hospital due to multisystem organ failure after device change out. The mean hospital length of stay for the remaining 19 patients was 20 ± 11 days. B-type natriuretic peptide levels are presented in Figure 1. The left ventricular end diastolic dimension (LVEDD) [preoperative mean ± SD 68.6 ± 11.2 mm; postoperative first week 56.5 ± 16.1 mm (p < 0.001); first month 56.9 ± 14.2 mm (p < 0.001)] and the degree of mitral regurgitation (MR) (0, none to trivial; 1, mild; 2, moderate; 3, moderate to severe; 4 severe) [preoperative mean ± SD 3 ± 1; postoperative first week 0.7 ± 0.8 (p < 0.001); first month 0.9 ± 1.1 (p < 0.001)] diminished significantly within the first week, whereas the right ventricular systolic pressure (RVSP) [preoperative mean ± SD 59 ± 13 mm Hg; postoperative first week 43 ± 11 mm Hg; first month 37 ± 10 mm Hg (p < 0.001)] diminished rather slowly over the several weeks after LVAD implantation. Interestingly, the right ventricular (RV) dysfunction persisted on serial echocardiography, despite the effective unloading of the left ventricle with LVAD. Patients' plasma BNP levels decreased over time as demonstrated in Figure 1. The overall trend of decrease of BNP <500 pg/ml was readily noticeable in the majority of patients who had an uncomplicated postoperative course (Figure 2). However, a few patients developed BNP patterns that defied the general trend of decrease (Figure 3). The rise in BNP levels corresponded with a state of compromised hemodynamic support due to various problems such as LVAD malfunction, tamponade, or worsening right ventricular dysfunction (Table 2). The corrective measures needed to restore the effective circulation included evacuation of hematoma, enhanced inotropic support for the failing right ventricle, and/or LVAD change out for thrombi formation at the impeller. Once the adequate forward flow was re-established, the BNP levels decreased again as expected.

Table 1

Table 1

Table 2

Table 2

Figure 1.

Figure 1.

Figure 2.

Figure 2.

Figure 3.

Figure 3.

Overall, the plasma BNP levels seem to reflect the pump speed settings of LVAD. The correlation was statistically significant for the entire group (Supplemental digital content, Table 1 at http://links.lww.com/ASAIO/A3) and individual patients while BNP levels were investigated in different pump speeds (see Supplemental Digital content, Figure 1 at http://links.lww.com/ASAIO/A1 and Figure 2 at http://links.lww.com/ASAIO/A2). The serum creatinine levels also reflected the various pump speed settings, but the correlation was not as significant as that of BNP.

B-type natriuretic peptide levels after a change in RPM (without including other pertinent data that might affect BNP levels such as transfusion, dialysis, inotropic medication use, or existence of RV failure) were analyzed. B-type natriuretic peptide values were significantly less than before at first 2 days after RPM increase (prechange BNP, 1,420 ± 1,125 pg/ml; postchange day 1, 1,168 ± 918 pg/ml; postchange day 2, 912 ± 599 pg/ml; p = 0.028). Significant increase of BNP was only observed the following first day when RPM was decreased (p = 0.008) but not after (prechange BNP, 1,046 ± 927 pg/ml; postchange day 1, 1,471 ± 1,177 pg/ml; postchange day 2, 1,153 ± 930 pg/ml; p = 0.13).

Back to Top | Article Outline

Long-Term Follow-Up

In the BTT group involving nine patients, four underwent cardiac transplantation. Two patients, who were weaned from LVAD successfully after the myocardial recovery, had a rapid reduction of their BNP values during their postoperative period. The values were reduced to <200 pg/ml 1 month after LVAD implantation. The patient with postpartum cardiomyopathy was successfully weaned after 4 months of VAD support. The other patient with dilated cardiomyopathy developed a device failure due to a broken driveline. His device was defunctionalized because his heart was recovered. One patient who had been supported on LVAD for a year died suddenly at home. The remaining patients are still awaiting heart transplantation on LVAD support.

There were two late deaths in the DT group. One died from an iatrogenic noncardiac cause whereas the other died from a hemorrhagic stroke after cerebral embolism.

Back to Top | Article Outline

Discussion

The BNP is a well-studied biological marker in patients with heart failure.8,9 It is well known that a higher level of BNP is associated with a worse prognosis in patients with advanced heart failure. However, the natural history and the physiology of patients with advanced heart failure can be altered greatly with LVAD therapy. After implantation of an LVAD, the LV is unloaded and the cardiac output improves immediately.10 During this period of rapid hemodynamic change, we followed serial BNP levels to see whether they would reflect the physiological improvement associated with a successful VAD support. The highly elevated preoperative BNP levels decreased steadily over the next 3 months, provided there were no problems with hemodynamic support rendered by LVAD. When patients encountered the situations of inadequate circulatory support because of problems such as acute LVAD failure, thrombus on impeller, tamponade, and/or profound RV dysfunction, their BNP levels increased noticeably. Then, it was reassuring again to witness the return of the decreasing trend of BNP levels following the corrective measures to address the problems that contributed to the inadequate circulatory support.

The need to follow a physiological marker such as BNP is more compelling in the era of continuous-flow LVADs. The earlier LVADs were based on the volume displacement technology and provided given stroke volumes and rates similar to the native heart. The usual clinical parameters such as blood pressures and pump rates could be followed to assess a patient's condition. However, with the newer generation LVADs, the blood is propelled in a continuous manner. Patients lack palpable pulse rates and pressures but only pulses measurable by Doppler. Following the trend of normalization of values of a physiological marker such as BNP may assure clinicians that LVAD support is adequate in providing forward flow and unloading of the LV. Furthermore, BNP levels may serve as a helpful physiological feedback parameter in adjusting the pump speed for an optimal setting in each patient. For example, a higher pump speed would be needed in patients with systemic hypertension or with an aortic insufficiency to compensate for a regurgitant flow. Although the echocardiography would be very helpful in understanding the degree of LV unloading, it makes no inference to the physiological impact of such unloading. It would be reassuring to clinicians to witness normalization of BNP levels in these patients.

The half-life of BNP in the circulation is quite short, with a mean length of 20 minutes. B-type natriuretic peptide accurately reflects current ventricular status. Generally, a BNP concentration <100 pg/ml would make the diagnosis of heart failure unlikely. There is a diagnostic “gray area,” often defined as between 100 and 500 pg/ml, for which the test is considered inconclusive.11,12 Values >500 pg/ml are generally considered to be positive in CHF. Few data are available on serial measurements of BNP in advanced chronic heart failure. In the Valsartan Heart Failure Trial (ValHeFT), those patients with the greatest reduction in their BNP concentration had the lowest mortality over the course of the study.13 We observed similarly uneventful hospital course and shorter hospital stay in our patients whose BNP levels lowered <500 pg/ml in couple of weeks after LVAD therapy.

There are several interventions or clinical conditions that may have occurred simultaneously with the modification of the device speed such as adjusting medical therapy (i.e., diuretics, neurohormonal blockade, and right ventricular dysfunction) that could have a similar effect on BNP. To lessen the degree of these confounders, we analyzed BNP levels just after RPM change (different time points between patients) and observed how quickly and effectively levels were decreased after RPM is increased. These were more prominent in 2 days after change. The similar magnitude of change was only observed first day when RPM was decreased. We thought that the need for increase in RPM was related to well being in hemodynamics, echocardiographic findings of LV filling, or aortic valve opening, which all will be reflected as a decrease in BNP by better LV unloading. On contrary, a need for RPM decrement would be a suction effect and/or septal shifting causing RV failure, which all may reflect as BNP decrease with better hemodynamics re-established even when speed is lowered.

The correlation of BNP with RPM was statistically apparent with the HeartMate II but not Jarvik 2000. Various factors such as a wider range of adjustable RPM, a more effective unloading of the LV, and a more complete hemodynamic support with the Heartmate II over Jarvik 2000 might have yielded the better correlation with BNP. We also encountered more episodes of troublesome hemodynamic support and pump failures with Jarvik 2000. This might have contributed to the less tight correlation between BNP and RPM in patients with Jarvik 2000.

The continuous-flow physiology created with LVAD support seems to be well tolerated clinically. Yet, our understanding is poor in this new physiology as to what constitutes the optimal LVAD support setting. How much unloading of the LV is ideal? What are the appropriate mean blood and pulse pressures? What guides us in setting the RPM? Adjusting the RPM of continuous LVAD based on a physiological feedback biomarker such as BNP might be helpful in managing these patients with this entirely new physiology. Clearly, we are facing a growing number of patients supported on continuous-flow LVAD, and we must be eager to learn this new uncharted physiology and its implication for our patients.

Back to Top | Article Outline

Limitations

The limitations include those inherent to an observational study. We had a different variety of patients in etiology of heart failure, including ischemic cardiomyopathy, which reported to have weak release of BNP due to necrosis and myocardial stiffness. Nevertheless, we observed satisfactory post-LVAD BNP decrease in this etiology.

Furthermore, there are several clinical conditions that may have occurred simultaneously with the modification of the device speed such as adjusting medical therapy (i.e., diuretics, dialysis, transfusions, and neurohormonal blockade right ventricular dysfunction) that could have a similar effect on BNP. All those confounders' magnitude in BNP for each patient for a given time is difficult to estimate. Addition to that one level of RPM may mean a different level of support and unloading between patients reflecting different BNP levels. We investigated each patient separately grouping altered speed levels in each other in different time points trying to minimize effect of such confounders. Also, whenever a change was present in RPM, BNP levels were analyzed, which might lessen the effect of these confounders.

Patients with no alteration with RPM or patients with less alteration duration at that RPM level generated none or very few BNP levels to perform sufficient correlation analysis. Related to prospective nature of the study, neither of these patients, other than clinical indications of primary responsible service, got alterations with RPM to increase the power of this study to have more correlation. Three patients with no alteration of RPM were included only in overall BNP level follow- up and correlation of BNP with creatinine but not in correlation of BNP with RPM. Including larger numbers of study patients or investigating with alternating RPM in animal studies may provide powerful statistical analysis to support results of this study.

Back to Top | Article Outline

Conclusion

There is a relationship between adjustable speed of continuous-flow LVAD and BNP levels especially in patients who had a stable clinical course after LVAD implantation. Serial BNP measurements are helpful to monitor and optimize left ventricular unloading on axial-flow pump support.

Back to Top | Article Outline

References

1. Mäntymaa P, Vuolteenaho O, Marttila M, Ruskoaho H: Atrial stretch induces rapid increase in brain natriuretic peptide but not in atrial natriuretic peptide gene expression in vitro. Endocrinology 133: 1470–1473, 1993.
2. Levin ER, Gardner DG, Samson WK: Natriuretic peptides. N Engl J Med 339: 321–328, 1998.
3. Yasue H, Yoshimura M, Sumida H, et al: Localization and mechanism of secretion of B-type natriuretic peptide in comparison with those of A-type natriuretic peptide in normal subjects and patients with heart failure. Circulation 90: 195-203, 1994.
4. Sodian R, Loebe M, Schmitt C, et al: Decreased plasma concentration of brain natriuretic peptide as a potential indicator of cardiac recovery in patients supported by mechanical circulatory assist systems. J Am Coll Cardiol 38: 1942–1949, 2001.
5. Xydas S, Rosen RS, Ng C, et al: Mechanical unloading leads to echocardiographic, electrocardiographic, neurohormonal, and histologic recovery. J Heart Lung Transplant 25: 7-15, 2006.
6. Loebe M, Muller J, Hetzer R: Ventricular assistance for recovery of cardiac failure. Curr Opin Cardiol 14: 234–248, 1999.
7. Khan T, Delgado RM, Radovancevic B, et al: Dobutamine stress echocardiography predicts myocardial improvement in patients supported by left ventricular assist devices (LVADs): Hemodynamic and histologic evidence of improvement before LVAD explantation. J Heart Lung Transplant 22: 137–146, 2003.
8. Potapov EV, Hennig F, Wagner FD, et al: Natriuretic peptides and E-selectin as predictors of acute deterioration in patients with inotrope-dependent heart failure. Eur J Cardiothorac Surg 27: 899–905, 2005.
9. Troughton RW, Prior DL, Pereira JJ, et al: Plasma B-type natriuretic peptide levels in systolic heart failure: Importance of left ventricular diastolic function and right ventricular systolic function. J Am Coll Cardiol 43: 416–422, 2004.
10. Madigan JD, Barbone A, Choudhri AF, et al: Time course of reverse remodeling of the left ventricle during support with a left ventricular assist device. J Thorac Cardiovasc Surg 121: 902–908, 2001.
11. Strunk A, Bhalla V, Clopton P, et al: Impact of the history of congestive heart failure on the utility of B-type natriuretic peptide in the emergency diagnosis of heart failure: Results from the Breathing Not Properly Multinational Study. Am J Med 119: 69.e1–69.e11, 2006.
12. Brenden CK, Hollander JE, Guss D, et al: Gray zone BNP levels in heart failure patients in the emergency department: Results from the Rapid Emergency Department Heart Failure Outpatient Trial (REDHOT) multicenter study. Am Heart J 151: 1006–1011, 2006.
13. Anand IS, Fisher LD, Chiang YT, et al: Changes in brain natriuretic peptide and norepinephrine over time and mortality and morbidity in the Valsartan Heart Failure Trial (Val-HeFT). Circulation 107: 1278–1283, 2003.

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2010 by the American Society for Artificial Internal Organs