http://www.jhltonline.org

ORIGINAL CLINICAL SCIENCE

Macrovascular and microvascular function after implantation of left ventricular assist devices in end-stage heart failure: Role of microparticles Roberto Sansone, MD,a Beate Stanske, MD,a Stefanie Keymel, MD,a Dominik Schuler, MD,a Patrick Horn, MD,a Diyar Saeed, MD,b Udo Boeken, MD,b Ralf Westenfeld, MD,a Artur Lichtenberg, MD,b,c Malte Kelm, MD,a,c and Christian Heiss, MDa From the aDivision of Cardiology, Pulmonology, and Vascular Medicine; bDepartment of Cardiovascular Surgery; and the cCardiovascular Research Institute, Heinrich Heine University, Medical Faculty, University of Düsseldorf, Medical Faculty, Düsseldorf, Germany.

KEYWORDS: left ventricular assist device; endothelial function; microvascular perfusion; microparticles

BACKGROUND: The hemodynamic vascular consequences of implanting left ventricular assist devices (LVADs) have not been studied in detail. We investigated the effect of LVAD implantation compared with heart transplant (HTx) on microvascular and macrovascular function in patients with end-stage heart failure and evaluated whether microparticles may play a role in LVAD-related endothelial dysfunction. METHODS: Vascular function was assessed in patients with end-stage heart failure awaiting HTx, patients who had undergone implantation of a continuous-flow centrifugal LVAD, and patients who had already received a HTx. Macrovascular function was measured by flow-mediated vasodilation (FMD) using high-resolution ultrasound of the brachial artery. Microvascular function was assessed in the forearm during reactive hyperemia using laser Doppler perfusion imaging and pulsed wave Doppler. Age-matched patients without heart failure and without coronary artery disease (CAD) (healthy control subjects) and patients with stable CAD served as control subjects. Circulating red blood cell (CD253þ), leukocyte (CD45þ), platelet (CD31þ/CD41þ), and endothelial cell (CD31þ/CD41, CD62eþ, CD144þ) microparticles were determined by flow cytometry and free hemoglobin by enzyme-linked immunosorbent assay. RESULTS: FMD and microvascular function were significantly impaired in patients with end-stage heart failure compared with healthy control subjects and patients with stable CAD. LVAD implantation led to recovery of microvascular function, but not FMD. In parallel, increased free hemoglobin was observed along with red and white cell microparticles and endothelial and platelet microparticles. This finding indicates destruction of blood cells with release of hemoglobin and activation of endothelial cells. HTx and LVAD implantation led to similar improvements in microvascular function. FMD increased and microparticle levels decreased in patients with HTx, whereas shear stress during reactive hyperemia was similar in patients with LVADs and patients with HTx. CONCLUSIONS: Our data suggest that LVAD support leads to significant improvements in microvascular perfusion and hemodynamics. However, destruction of blood cells may

Reprint requests: Christian Heiss, MD, Division of Cardiology, Pulmonology, and Vascular Medicine, Heinrich Heine University Düsseldorf, Medical Faculty, Moorenstrasse 5, 40225 Düsseldorf, Germany. Telephone: þ49-211-810-8753. Fax: þ49-211-811-8812. E-mail address: [email protected] 1053-2498/$ - see front matter r 2015 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2015.03.004

2

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]] contribute to residual endothelial dysfunction potentially by increasing nitric oxide scavenging capacity. J Heart Lung Transplant ]]]];]:]]]–]]] r 2015 International Society for Heart and Lung Transplantation. All rights reserved.

Left ventricular assist devices (LVADs) are considered a vital therapeutic option to assist temporarily or permanently the failing circulation in patients with end-stage heart failure as a bridge to heart transplant (HTx) or in elderly patients as destination treatment.1 Several studies demonstrated that continuous flow LVADs improve morbidity and mortality in critically ill patients awaiting HTx, while reducing adverse events.2,3 Most patients with a continuous flow system exhibited significantly increased daily functional capabilities compared with before implantation.4 Nevertheless, cardiovascular mortality and morbidity are still greater than after HTx. In this context, the impact of LVADs on arterial endothelial function has not been studied in detail. Heart failure is associated with endothelial dysfunction.5 Structural maintenance of arteries is coupled to blood flow to a significant degree by shear stress-dependent endothelial nitric oxide (NO) production. In the absence of endothelial nitric oxide synthase (eNOS), arteries lose their ability to couple blood flow demands to vascular structural remodeling.6 Decreased eNOS-dependent vasodilation, as measured by flow-mediated vasodilation (FMD), has been linked to increased progression of early atherosclerosis and poor cardiovascular outcome.7 Several mechanisms lead to impaired NO bioavailability. Free NO is scavenged rapidly by cell-free hemoglobin.8 Hemolysis is a relevant sequela of several LVAD types, especially axial LVADs.9 Destruction of red blood cells (RBCs) is the result of wall shear stress, flow acceleration, and interaction with artificial surfaces10 and leads to release of free hemoglobin and RBC microparticles, which are defined by the expression of CD235 antigen. Increased levels of RBC microparticles have been found in patients with sickle cell disease11 and β-thalassemia major,12 which are also characterized by hemolysis. However, the role of RBC microparticles in patients with LVADs is unknown. Microparticles are membrane particles with a diameter o1 μm that are thought to be shed from endothelial cells and various blood cells, including platelets, leukocytes, and erythrocytes, and released in the circulation.13–15 The microparticles in the circulation constitute a heterogeneous population of different cellular origins, numbers, size, and antigenic composition. Proposed mechanisms of microparticle generation include apoptosis and cellular activation by cytokine.13–16 NO synthesis and shear stress are key inhibitors of microparticle generation in endothelial cells.16 RBC microparticles may be released during hemolysis.17 Although microparticles circulate in blood from healthy individuals, their numbers are increased in blood samples from patients several cardiovascular diseases and conditions that predispose to cardiovascular disease,13–15 and they are associated with adverse outcomes in patients with ischemic chronic heart failure18 and in patients after LVAD

implantation.19 The literature suggests that the number of circulating microparticles may be a marker of endothelial activation or damage and platelet activation.13,15 In addition, it was appreciated that microparticles harbor numerous membrane and cytoplasmic proteins from the cells from which they originate20–22 and may play a role as a disseminated storage pool of bioactive effectors in intercellular communication mediating effects in cardiovascular physiology and pathophysiology.13–15,23,24 Endotheliumderived microparticles of healthy subjects may carry a functional eNOS that is associated with better endotheliumdependent vasodilation,25 whereas RBC microparticles appear to be potent NO scavengers owing to their hemoglobin content and parallel release of free hemoglobin during hemolysis.26 It is unknown whether microparticle generation, potentially through hemolysis,27 after LVAD implantation may affect vascular function. We investigated the effect of LVAD implantation compared with HTx on microvascular and macrovascular function in patients with end-stage heart failure and evaluated whether microparticles may play a role in LVAD-related endothelial dysfunction.

Methods Study subjects and protocol The study protocol was approved by the institutional review board of Heinrich-Heine University, and all subjects gave written informed consent (ClinicalTrials.gov Identifier: NCT02174133). Patients included in the study were screened and recruited from the patients routinely presenting in the specialized clinic within the Düsseldorf Heart Failure Program and were awaiting HTx (endstage heart failure; status highly urgent was excluded from the study), had already received a Htx transplant, or were provided with a LVAD as a bridge to transplantation according to the definitions in the current guidelines.28 Clinical data measurements were taken at 3 months after HTx or LVAD implantation. Patients with pump thrombosis were routinely excluded clinically and according to data retrieval from LVAD. Subjects for the 2 control groups were recruited from the general cardiology outpatient clinic. We included patients with stable coronary artery disease (CAD) and normal systolic left ventricular function (left ventricular ejection fraction 460%) and healthy male subjects without signs or symptoms of cardiovascular disease who presented for a routine checkup. During the screening interview, control subjects were excluded if a physician had ever told them that they had cardiovascular disease, including coronary, lower limb, or carotid artery disease; they had experienced a myocardial infarction; they were taking any medication indicative of cardiovascular disease, including statins, daily aspirin, anti-diabetic medication (insulin, metformin), or blood pressure–lowering medication (angiotensinconverting enzyme inhibitor/angiotensin receptor blocker, beta blocker, diuretic, calcium channel blocker); or had symptoms

Sansone et al.

Micro- and macrovascular function after LVAD implantation

indicative of cardiovascular disease, including dyspnea, chest pain, palpitations, syncope, or claudication.

LVAD All subjects who had received a LVAD were implanted with a HeartWare system (HeartWare, Inc). Briefly, the pump was implanted pericardially with an integrated inflow cannula inserted into the apex of the heart. The integrated inflow cannula contains a centrifugal flow pump with a contactless hydrodynamic rotor (1,800–4,500 rpm) that delivers continuous blood flow (up to 10 liters/min) via an outflow graft tube that is anastomosed to the ascending aorta.

FMD Brachial artery FMD was measured by ultrasound (10 MHz transducer; Vivid i; GE Healthcare) in combination with an automated analysis system (Brachial Analyzer; Medical Imaging Applications, Iowa City, IA) in a 211C temperature-controlled room after 15 minutes of rest in the supine position.29 A forearm blood pressure cuff was placed distal to the cubital fossa and inflated to 250 mm Hg for 5 minutes. Diameter and angle-corrected Doppler flow velocity readings were measured at baseline and immediately after cuff deflation, at 20, 40, 60, 80, 100, and 120 sec. FMD was expressed relative to baseline diameters as (diametermaximum  diameterbaseline)/diameterbaseline. Wall shear stress was calculated as 8 * m * mean flow velocity/mean diameter, where blood viscosity (m) was assumed to be constant at 0.035 dyne/sec * cm2. Because of contraindication of sublingual nitroglycerin in hypotensive patients, we were unable to test a nitroglycerin-mediated vasodilation.

Hemodynamic monitoring Heart failure symptoms were classified using the New York Heart Association classification. Resting two-dimensional echocardiography was performed at baseline.30 Biplane left ventricular ejection fraction assessment according to Simpsonʼs rule was obtained for each patient in the apical 2-chamber and 4-chamber view.31 Volumetric parameters were acquired by biplane summation as reliably feasible in the individual patient. Mean arterial pressure was calculated as diastolic blood pressure þ ([systolic blood pressure  diastolic blood pressure]/3).

Microvascular function assessed by non-invasive laser Doppler imaging All investigations were performed using a scanning laser Doppler perfusion imager (PeriScan PIM 3 System; Perimed AB, Stockholm, Sweden). After 15 minutes of rest, the first scan was started. To avoid moving artifacts, the arm selected for measurements was immobilized using a vacuum pillow containing polyurethane beads, which molds to the shape of the arm (Germa AB; Kristianstad, Sweden). The laser beam was positioned 15 cm above the forearm scanning a field of 200 mm2 (region of interest ¼ 8  8 pixels; 3 sec per scan) on the volar site of the forearm. Microvascular reactivity was assessed during post-occlusive reactive hyperemia following 5 minutes of forearm occlusion. Following the baseline perfusion measurements (1 minute; 20 images), a blood pressure cuff located at the distal upper arm was inflated to supra-systolic levels over 5 minutes.

3

After the cuff release, the microvascular response on reactive hyperemia was recorded. Data acquisition and analysis were performed by LDPIwin Software (Perimed AB) processing the perfusion as numerical values and color-coded images. Maximum perfusion, amplitude of perfusion (maximum perfusion  baseline perfusion), ratio (maximum perfusion/baseline perfusion), percentage increase ([maximum perfusion  baseline perfusion]/baseline perfusion  100), time to peak response, and area under curve were evaluated without subtracting biologic zero from the data.

Characterization of microparticle sub-populations by flow cytometry Citrated blood (6 ml) was drawn from the cubital vein from control subjects and patients with LVADs and processed within 2 hours. Platelet-rich plasma was obtained by centrifugation of whole blood at 300g over 15 minutes at room temperature. Platelet-free plasma was obtained by 2 successive centrifugations of platelet-rich plasma at 10,000g for 5 minutes at room temperature. Microparticle pellets and microparticle-free plasma samples were obtained by ultracentrifugation of the platelet-free plasma at 30,000g for 90 minutes at 41C. The protein concentration in the plasma of all blood donors did not differ significantly. Microparticle sub-populations were discriminated by flow cytometry according to the expression of established surface antigens as described previously.25 Briefly, samples were incubated for 30 minutes with fluorochrome-labeled antibodies or matching isotype controls and analyzed in a FACSCanto II flow cytometer (Becton Dickinson, Heidelberg, Germany). Microbead standards (1.0 μm) were used to define microparticles as o1 μm in diameter. The microparticle sub-populations were defined as follows: CD31þ/ CD41þ microparticles as platelet-derived microparticles; CD62eþ, CD144þ, or CD31þ/CD41 microparticles as endothelium-derived microparticles; CD235þ microparticles as erythrocyte-derived microparticles (RBC microparticles); and CD45þ microparticles as leukocyte-derived (white blood cell) microparticles. The total number of microparticles was quantified with flow-count calibrator beads (20 μl).

Statistical methods Results are expressed as mean ⫾ SD. Baseline data represent data of first visit. The primary test for an effect was 2-way repeated measure analysis of variance and Bonferroni post-hoc test, and confidence intervals for pair-wise comparisons were computed with IBM SPSS Statistics version 20 (IBM Corp). Correlations were Pearson r. p-values o 0.05 were regarded as statistically significant.

Results Baseline characteristics of study subjects See Figure 1 for study flow and Table 1 for characteristics of the study population.

LVAD implantation and HTx lead to improved microvascular function The studied groups did not differ with regard to baseline (resting) perfusion of the cutaneous microcirculation as

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

4

Figure 1 CONSORT (Consolidated Standards of Reporting Trials) diagram. CAD, coronary artery disease; LVAD, left ventricular assist device; HTx, heart transplant.

measured by LDPI between healthy controls, patients with stable CAD, and patients with end-stage heart failure (37 ⫾ 6 PU vs 36 ⫾ 17 PU vs 29 ⫾ 10 PU). Although the capacity to increase perfusion in response to forearm ischemia was already impaired in patients with stable CAD compared with healthy controls, it was significantly further impaired in patients with end-stage heart failure. Typical LDPI curves recorded during reactive hyperemia are depicted in Figure 2. Patients with end-stage heart failure exhibited significant reductions in all LDPI parameters reflecting Table 1

different components of microcirculatory function (maximal perfusion, amplitude, and area under the curve [AUC]). Although resting perfusion was unaffected by LVAD implantation (39 ⫾ 7 PU, analysis of variance p ¼ 0.1), it restored the capacity of the microcirculation to increase perfusion in response to ischemia compared with patients with end-stage heart failure in all parameters: maximal perfusion (73 ⫾ 16 PU vs 145 ⫾ 39 PU, p o 0.011), amplitude (45 ⫾ 18 PU vs 125 ⫾ 30 PU, p o 0.004), and AUC (809 ⫾ 451 PU vs 2,117 ⫾ 312 PU, p o 0.001).

Characteristics of Study Population

Number of male patients Age (years) BMI (kg/m2) eGFR (ml/min * 1.73 m2) Dilated cardiomyopathy (%) Ischemic cardiomyopathy (%) Myocarditis (%) Hypertension (%) Diabetes mellitus (%) Dyslipidemia (%) NYHA (I, II, III, IV) Hemodynamics CO (liters/min) HR (beats/min) MAP (mm Hg) Hemoglobin (mg/dl) Cell-free hemoglobin (mg/dl) AST (mg/dl) ALT (mg/dl) LDH (U/liter) CRP (mg/dl) BNP (ng/liter) HbA1c (%) Total cholesterol (mg/dl)

Healthy

CAD

Heart failure

LVAD

HTx

10 60⫾7 25.7⫾1.9 80⫾8 0 0 0 0 0 80 10/0/0/0

10 57⫾6 28.2⫾3.6 77⫾4 0 0 0 60 40 100 7/3/0/0

10 51⫾9 25.7⫾4.7 67⫾23 30 50 20 50 40 40 0/0/5/5

14 61⫾9 26.2⫾5.7 73⫾19 14 78 8 60 15 30 1/11/2/0

8 49⫾12 24.4⫾3.7 74⫾5 25 50 25 14 14 28 6/2/0/0

5.1⫾0.2 56⫾6 95⫾13 14.4⫾0.5 27⫾4 27⫾10 26⫾10 156⫾24 0.6⫾0.6 18⫾11 5.4⫾0.4 214⫾19

5.1⫾0.5 69⫾12 106⫾5 13.6⫾1.2 50⫾16 26⫾7 34⫾10 171⫾13 0.8⫾0.7 30⫾13 5.8⫾0.2 170⫾11

2.1⫾0.6 96⫾12 69⫾4 12.3⫾1.7 100⫾29 77⫾88 134⫾145 295⫾105 0.9⫾1.0 9,877⫾11,544 6.2⫾0.6 171⫾70

4.6⫾0.2 73⫾8 76⫾10 12.9⫾1.0 179⫾74 21⫾10 27⫾10 231⫾82 0.4⫾0.2 656⫾700 6.1⫾0.2 174⫾52

5.0⫾0.2 68⫾4 92⫾7 12.8⫾1.1 47⫾17 22⫾12 26⫾18 163⫾18 0.3⫾0.4 65⫾20 5.7⫾0.3 199⫾37

p-value (ANOVA) 0.023 0.692 0.590

0.001 0.001 0.692 0.090 0.001 0.001 0.161 0.072 0.301 0.006 0.447 0.006

Values are presented as mean ⫾ SD unless otherwise noted. ALT, alanine aminotransferase; ANOVA, analysis of variance; AST, aspartate aminotransferase; BMI, body mass index; BNP, B-type natriuretic peptide; CAD, coronary artery disease; CO, cardiac output; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; HbA1c, glycosylated hemoglobin; HR, heart rate; HTx, heart transplant; LDH, lactate dehydrogenase; LVAD, left ventricular assist device; MAP, mean arterial pressure; NYHA, New York Heart Association.

Sansone et al.

Micro- and macrovascular function after LVAD implantation

5

Figure 2 Microcirculation measurements as laser Doppler perfusion imaging (LDPI) of the cutaneous microcirculation. Scanning LDPI in healthy control subjects (green), patients with stable coronary artery disease (CAD) (yellow), patients with end-stage heart failure (red), patients 3 months after receiving a left ventricular assist device (LVAD) (orange), and patients after receiving a heart transplant (HTx) (blue). (A) Typical post-occlusion reactive hyperemia response curve assessed by LDPI in a healthy control subject (green) and a patient with endstage heart failure (red). (B–E) Basal perfusion, area under the post-occlusion reactive hyperemia curve (AUC), maximal perfusion, and amplitude. #p o 0.05 vs healthy control subjects. *p o 0.05 vs end-stage heart failure patients.

Similar to LVAD, HTx did not change baseline perfusion (46 ⫾ 11 PU, analysis of variance p ¼ 0.1), but was accompanied by improvement of microcirculatory function compared with patients with end-stage heart failure: maximal perfusion (73 ⫾ 16 PU vs 172 ⫾ 42 PU, p ¼ 0.001), amplitude (45 ⫾ 18 PU vs 106 ⫾ 41 PU,

p o 0.019), and AUC (809 ⫾ 451 PU vs 2,248 ⫾ 582 PU, p o 0.001). Microcirculatory functional parameters did not differ significantly between patients with LVAD and patients with HTx. Qualitatively similar results were obtained with pulsed wave Doppler recorded along with FMD measurements (see further on).

6

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

HTx but not LVAD improves conduit artery endothelial function FMD was significantly decreased in patients with end-stage heart failure compared with healthy control subjects and control subjects with stable CAD demonstrating decreased functional capacity of conduit arteries to dilate in response to an increase in shear stress—endothelial dysfunction. As depicted in Figure 3, HTx led to an improvement in FMD compared with patients with end-stage heart failure (4.3 ⫾ 0.6% vs 2.6 ⫾ 0.6%, p o 0.001). There was no significant difference between patients with stable CAD and patients after HTx (4.1 ⫾ 0.2% vs 4.3 ⫾ 0.6%, p ¼ 1.000). FMD did not increase after LVAD implantation and remained impaired on the level of patients with end-stage heart failure (2.1 ⫾ 0.6% vs 2.6 ⫾ 0.6%, p ¼ 0.938). The baseline diameter of the brachial artery did not differ

between groups and remained unaffected by either intervention. To evaluate whether residual impairments in FMD as seen in the LVAD group may be due to a lowered shear stress stimulus during reactive hyperemia, we calculated wall shear stress in all groups (Figure 4). Our results qualitatively confirmed the above-described LDPI results and showed that wall shear stress was slightly, yet significantly, decreased in patients with stable CAD and severely lowered in patients with end-stage heart failure along with a decreased flow response after 5 minutes of forearm occlusion. There was no difference between wall shear stress during reactive hyperemia between LVAD and HTx. However, B-type natriuretic peptide (BNP) values were significantly elevated in patients who received LVADs compared with control subjects and patients after HTx, and BNP values inversely correlated with FMD (r ¼ 0.43, 0.035).

Figure 3 Endothelial function measurements as flow-mediated vasodilation (FMD) of the brachial artery. FMD in healthy control subjects (green), patients with stable coronary artery disease (CAD) (yellow), patients with end-stage heart failure (red), patients 3 months after receiving a left ventricular assist device (LVAD) (orange), and patients after receiving a heart transplant (HTx) (blue). (A) Impaired FMD in CAD patients, end-stage heart failure patients, and patients after LVAD compared with control subjects and recovery of FMD in patients after HTx. (B) No differences in baseline diameter of brachial artery. #p o 0.05 vs healthy control subjects. *p o 0.05 vs end-stage heart failure patients. (C) Scatterplot of free cell hemoglobin (fHb) and FMD (r ¼ 0.79; p ¼ 0.001).

Sansone et al.

Micro- and macrovascular function after LVAD implantation

7

Figure 4 Measurements of baseline and reactive hyperemic responses at the levels of brachial artery blood flow velocity (dotted lines), wall shear stress (WSS) (red lines), and diameter (black line), as expressed by flow-mediated vasodilation (FMD). Compared with healthy control subjects (A), patients with stable coronary artery disease (CAD) (B) exhibit slightly but significantly lowered peak flow velocity and WSS at the onset of reactive forearm hyperemia after 5 minutes of forearm occlusion and lowered peak FMD values. These impairments are significantly more pronounced in patients with end-stage heart failure (C). In patients after left ventricular assist device (LVAD) implantation (D) and heart transplant (HTx) (E), flow velocity and WSS were similarly improved to values of healthy control subjects (compare with Figure 2); FMD improved only to values of CAD patients but remained impaired in LVAD patients (compare with Figure 3). Baseline values did not differ between groups. *p o 0.05 vs respective time point in healthy control subjects. #p o 0.05 vs respective time point in CAD patients.

8

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

Evidence for destruction of RBCs and free hemoglobin after LVAD implantation: Relation to endothelial dysfunction LVADs may induce hemolysis by exerting high mechanical forces on circulating blood cells. Implantation of a LVAD led to increased levels of RBC microparticles and free hemoglobin suggesting increased RBC destruction and hemoglobin release. This suggestion was supported by a significant correlation between lactate dehydrogenase (LDH) and free hemoglobin across all groups (r ¼ 0.42; p ¼ 0.04). CD235þ microparticles were significantly elevated in patients with LVAD compared with healthy subjects (2,039 ⫾ 974 Ev/μl vs 646 ⫾ 325 Ev/μl, p o 0.0001) and patients with stable CAD (2,039 ⫾ 974 Ev/μl vs 1,290 ⫾ 628 Ev/μl, p o 0.048). We also observed increased levels of microparticles derived from leukocytes (CD45þ) and platelets (CD31þ/CD41þ). We previously showed that free hemoglobin can impair endothelial function potentially by scavenging of endothelium-derived NO.32 We observed consistently a strong inverse univariate correlation between FMD and free hemoglobin (r ¼ 0.79, p ¼ 0.001) (Figure 5 and 6).

LVAD implantation and impaired endothelial integrity In addition to microparticles derived from circulating blood cells, we observed increased levels of microparticles of endothelial origin (CD31þ/CD41, CD144þ, CD62eþ) after LVAD implantation. This finding suggests that LVAD implantation had a significant effect on the endothelium and potentially disturbed endothelial integrity. CD62eþ, the leukocyte-endothelial cell adhesion molecule 2, is exclusively expressed on activated endothelial cells, suggesting that endothelial cell activation is associated with LVAD.

Discussion The main finding of the present study is that a LVAD can restore microvascular function and cardiac output to a similar degree as HTx. However, macrovascular endothelial function remained decreased; this was related to RBC microparticles and free hemoglobin. In parallel, we observed that not only were blood cell microparticles increased in LVAD patients, but also endothelium-derived microparticles pointing toward endothelial activation or apoptosis (Figure 6).

Figure 5 Microparticle sub-populations in platelet-free plasma were discriminated by flow cytometry analysis of platelet-free plasma according to the expression of membrane-specific antigens: endothelial cells, CD31þ/CD41 (A), CD62eþ (B), CD144þ (C), and platelets (CD31þ/CD41þ) (D); leukocytes (CD45þ) (E); and erythrocytes (CD253þ) (F). Values are mean ⫾ SEM.

Sansone et al.

Micro- and macrovascular function after LVAD implantation

9

Figure 6 Graphic abstract illustrating the proposed role of microparticles in the pathophysiology of left ventricular assist device (LVAD)–related endothelial dysfunction (left). Even in healthy people, microparticles that originate from red and white blood cells, platelets, and endothelial cells circulate in plasma. Constitutively expressed and shear stress–stimulated nitric oxide (NO) production occurs that modulates vascular tone but that also can inhibit endothelial microparticle release (right). Blood cells and platelets are exposed to mechanical forces that can destroy blood cells and activate platelets leading to increased blood cells and platelet microparticles. Free hemoglobin (fHb) and red blood cell microparticles released during hemolysis are potent NO scavengers that may curb NO-dependent vasodilation and inhibition of endothelial microparticle release. Parallel endothelial activation and lack of pulsatile flow are likely important contributing factors with a negative impact on endothelial functions, including flow-mediated vasodilation.

It has been shown that the failing circulation in chronic heart failure is associated with a decreased capacity of the peripheral microcirculation (microvascular dysfunction) early on, which may contribute to symptoms such as fatigue and heat intolerance.33–35 The etiology of microvascular dysfunction is still not completely understood but was shown to be associated with a low-grade inflammation.34 In the present study, we confirmed that a severe degree of microvascular dysfunction exists in patients with end-stage heart failure and that it was normalized after LVAD implantation. Because C-reactive protein values did not differ significantly between the groups, our data suggest that inflammation most likely does not account for microvascular dysfunction in end-stage heart failure. Rather, our data support that improvements in microvascular function correlate with the hemodynamic improvement in these patients. In 83 consecutive patients implanted with continuous flow LVADs, Hasin et al36 showed a dramatic improvement in renal function with LVAD support suggesting the importance of improving cardiac output and hemodynamic status for the physiology of complex multiple-organ syndromes. With the assumption that LDPI and Doppler flow reserve measurements are sensitive and reproducible methods for non-invasive assessment of microvascular perfusion during reactive hyperemia,37 we conclude that LVAD is capable of improving peripheral perfusion by restoration of sufficient circulation. Besides microvascular dysfunction, heart failure is associated with an impairment of conduit artery endothelial function (macrovascular dysfunction). In the present study, we observed significantly impaired flow-mediated vasodilation, an established marker of macrovascular function, in patients with end-stage heart failure. Similar to

microvascular dysfunction in heart failure, inflammation is known to play an important role in heart failure–associated macrovascular dysfunction.38 As discussed earlier, Creactive protein values did not differ significantly between groups suggesting also that the different FMD values between the groups cannot be primarily ascribed to inflammation. The degree of endothelial dysfunction is known to correlate with heart failure severity and functional capacity.39 Hambrecht et al40 demonstrated that recovery of endothelial function is a major determinant of cardiac outcome. After HTx, patients reached FMD values equivalent to patients with stable CAD and significantly improved FMD values compared with patients with end-stage heart failure. This observation may be related to the fact that about 50% of the HTx population received an organ for terminal ischemic cardiomyopathy secondary to atherosclerotic CAD, a condition that is associated with and even believed to be caused by impaired endothelial function with decreased eNOS activity.41,42 However, an interesting observation of our study is that macrovascular endothelial function improved in patients after HTx but not in patients after LVAD implantation. As depicted in Figure 4, LVAD implantation led to a restoration of microvascular vasodilator function with increased hyperemic blood flow and wall shear stress. However, FMD remained lower compared with patients after HTx. Potential explanations for residual macrovascular dysfunction in patients with LVADs include continuous blood flow or lack of pulsatile blood flow, hemolysis, submaximal hemodynamic support, and endothelial activation. In vitro and in vivo studies have shown that pulsatile blood flow is important for vascular homeostasis. Pulsatile shear stress and cyclic strain maintain endothelial survival,43

10

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

normal structure, and function, whereas stasis, turbulent flow, and local shear gradients activate endothelial cells and induce a proatherogenic state characterized by impaired expression of eNOS and NO bioavailability.44 In this context, Amir et al45 showed in vivo that pulsatile LVAD support is associated with a better FMD than with an axial continuous flow LVAD. A physiologic repetitive short-term increase in flow is an important contributing factor to improved peripheral vascular reactivity. Shechter et al46 reported the positive effect of arterial flow and pulsatility on endothelial function by external counterpulsation. Another factor potentially contributing to macrovascular endothelial dysfunction may be decreased NO availability secondary to increased scavenging by free hemoglobin and RBC microparticles. In clinical practice, overt hemolysis with LDH elevation 4350 U/liter is an important indicator of imminent LVAD pump thrombosis. The only marginally increased LDH values observed in our study together with a significant correlation with free hemoglobin in the absence of signs of pump thrombosis rather indicates low-grade hemolysis. Properly functioning LVADs always lead to reduced haptoglobin and hemopexin values and higher levels of free hemoglobin and LDH.9 However, in centrifugal LVAD, in particular with hydrodynamic bearings, hemolysis was an issue. More recent literature does not definitively support any difference in hemolysis between current axial and centrifugal continuous flow pumps.47 However, even low levels of free hemoglobin may have pathophysiologic consequences.32 Mechanistically, NO reacts with oxyhemoglobin in a rapid and irreversible reaction that produces nitrate and methemoglobin. Small amounts of cell-free hemoglobin and RBC microparticles are sufficient to offset endothelial NO production completely and result in endothelial dysfunction.26,48,49 These findings are in line with previous findings from our group showing that most NO scavenging activity of hemodialysis patient plasma is explained by the plasma cell-free hemoglobin/heme content.32 Several studies reported that NO reacts with RBC encapsulated hemoglobin much more slowly than when the hemoglobin is free in solution or plasma.50,51 In vivo, blood flow pushes RBCs to the center of the vessel, away from the NO-producing endothelium leading to a relative cell-free zone closed to the endothelium; this might contribute to reduced NO scavenging by RBCs.52 Nevertheless, in the case of LVAD implantation with consecutive hemolysis, this protection is abolished, and NO scavenging becomes more important as hemoglobin concentrations become higher, particularly in the cell free border zone close to the endothelium. In addition to release of cell-free hemoglobin, RBC breakdown leads to formation of hemoglobin-containing endothelial microparticles, as we have found in the patients with LVADs in the present study. Because of their small size, RBC microparticles were shown to be potent NO scavengers.17,26 In the present study, we showed a strong inverse correlation between cell-free hemoglobin and endothelial function as measured by FMD. Impaired endothelial function after LVAD implantation may be due to increased NO scavenging secondary to LVAD-associated hemolysis.

BNP not only may be a biomarker of heart failure severity, but it is also inversely related to endothelial function.53 Studies suggest that BNP possesses endothelial bioactivity that may be causally linked to heart failure– related endothelial dysfunction.54 In the present study, patients with LVADs exhibited significantly elevated BNP levels, which may account for residual endothelial dysfunction. It was previously shown that LVAD implantation leads to activation of endothelial and coagulation systems.55,56 As discussed earlier, endothelial activation may be due partly to the lack of pulsatile flow.43 Our data support this possibility by demonstrating increased concentrations of endotheliumderived (in particular, CD62eþ microparticles) and plateletderived microparticles in patients with LVADs suggesting activation of endothelial cells and platelets. Although the increased levels of platelet-derived microparticles may be explained by exposure to mechanical forces during LVAD passage leading to activation, the increased levels of endothelium-derived microparticles can be interpreted as markers of impaired endothelial function secondary to endothelial activation or apoptosis.13–16 This interpretation is in line with previously published studies suggesting that recipients of LVADs experience significant activation of endothelial and coagulation systems. The expression of intercellular adhesion molecule, E-selectin, and tissue factor on circulating endothelial cells strongly increased postoperatively and remained elevated at a lower level 180 days after LVAD implantation.56 In parallel, prothrombin fragments, D-dimers, and thrombin/anti-thrombin were elevated in these patients. Serum from patients with LVADs contains high levels of tumor necrosis factor-α, soluble tumor necrosis factor-α receptor, and interleukin-10. The incubation of serum from patients after HTx and patients after LVAD implantation with human pulmonary artery endothelial cells in vitro led to nuclear factor κB activation.55 Several limitations need to be taken into account when evaluating the present study. First, we did not follow individual patients, and the results presented reflect differences between groups in a cross-sectional observation. In patients with LVADs, blood flow was continuous and not pulsatile. This fact makes it difficult to distinguish definitively the effects of continuous flow or consequences of a lack of pulsatility from other pathophysiologic vascular factors (e.g., free hemoglobin). Pulsatile blood flow may be a prerequisite to allow shear stress to be transduced into a vasodilator response of conduit arteries. However, we previously demonstrated that free hemoglobin affects endothelial function in subjects with pulsatile blood flow.32 In conclusion, our data suggest that implantation of a LVAD leads to significant improvements in microvascular perfusion. However, destruction of blood cells may contribute to residual endothelial dysfunction potentially by increasing NO scavenging capacity (Figure 6). The present data underscore the potential need for developing novel biologic surfaces of implants, delineate the potential importance of pulsatile flow for vascular homeostasis, and point toward an important pathophysiologic role of endothelial dysfunction and NO deficit in the context of

Sansone et al.

Micro- and macrovascular function after LVAD implantation

implant thrombosis and inflammatory or immunomodulatory responses to implants. The biomarkers studied here may serve as important prognostic readouts for risk stratification allowing the development of novel devices and therapeutic strategies to improve further the long-term outcome of patients with LVADs and help identify patients who require early HTx or intensified therapy. Prospective long-term studies are needed.

Disclosure statement This study was funded by the Deutsche Forschungsgemeinschaft (Grant No. KE405/5-1 to M.K. and Grant No. GRK 1902 TP9 to M.K. and C.H.) and the Forschungskommission of the Medical Faculty of Heinrich-Heine University Düsseldorf (to C.H.). M.K. was supported by the Susanne Bunnenberg Stiftung at the Düsseldorf Heart Center.

References 1. Daneshmand MA, Rajagopal K, Lima B, et al. Left ventricular assist device destination therapy versus extended criteria cardiac transplant. Ann Thorac Surg 2010;89:1205-9: discussion 1210. 2. Miller LW, Pagani FD, Russell SD, et al. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med 2007; 357:885-96. 3. Slaughter MS, Rogers JG, Milano CA, et al. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009;361:2241-51. 4. Rogers JG, Aaronson KD, Boyle AJ, et al. Continuous flow left ventricular assist device improves functional capacity and quality of life of advanced heart failure patients. J Am Coll Cardiol 2010;55: 1826-34. 5. Chan SY, Mancini GB, Kuramoto L, Schulzer M, Frohlich J, Ignaszewski A. The prognostic importance of endothelial dysfunction and carotid atheroma burden in patients with coronary artery disease. J Am Coll Cardiol 2003;42:1037-43. 6. Heiss C, Rodriguez-Mateos A, Kelm M. Central role of eNOS in the maintenance of endothelial homeostasis. Antioxid Redox Signal 2014 Oct 20: [Epub ahead of print]. 7. Ras RT, Streppel MT, Draijer R, Zock PL. Flow-mediated dilation and cardiovascular risk prediction: a systematic review with meta-analysis. Int J Cardiol 2013;168:344-51. 8. Rother RP, Bell L, Hillmen P, Gladwin MT. The clinical sequelae of intravascular hemolysis and extracellular plasma hemoglobin: a novel mechanism of human disease. JAMA 2005;293:1653-62. 9. Heilmann C, Geisen U, Benk C, et al. Haemolysis in patients with ventricular assist devices: major differences between systems. Eur J Cardiothorac Surg 2009;36:580-4. 10. Yasuda T, Shimokasa K, Funakubo A, Higami T, Kawamura T, Fukui Y. An investigation of blood flow behavior and hemolysis in artificial organs. ASAIO J 2000;46:527-31. 11. Tantawy AA, Adly AA, Ismail EA, Habeeb NM, Farouk A. Circulating platelet and erythrocyte microparticles in young children and adolescents with sickle cell disease: relation to cardiovascular complications. Platelets 2013;24:605-14. 12. Tantawy AA, Adly AA, Ismail EA, Habeeb NM. Flow cytometric assessment of circulating platelet and erythrocytes microparticles in young thalassemia major patients: relation to pulmonary hypertension and aortic wall stiffness. Eur J Haematol 2013;90:508-18. 13. Amabile N, Rautou PE, Tedgui A, Boulanger CM. Microparticles: key protagonists in cardiovascular disorders. Semin Thromb Hemost 2010; 36:907-16. 14. Burnier L, Fontana P, Kwak BR, Angelillo-Scherrer A. Cell-derived microparticles in haemostasis and vascular medicine. Thromb Haemost 2009;101:439-51.

11

15. George FD. Microparticles in vascular diseases. Thromb Res 2008;122 (Suppl 1):S55-9. 16. Vion AC, Ramkhelawon B, Loyer X, et al. Shear stress regulates endothelial microparticle release. Circ Res 2013;112:1323-33. 17. Kim-Shapiro DB, Lee J, Gladwin MT. Storage lesion: role of red blood cell breakdown. Transfusion 2011;51:844-51. 18. Berezin AE, Kremzer AA, Samura TA, Martovitskaya YV. Circulating endothelial-derived apoptotic microparticles in the patients with ischemic symptomatic chronic heart failure: relevance of pro-inflammatory activation and outcomes. Int Cardiovasc Res J 2014;8:116-23. 19. Nascimbene A, Hernandez R, George JK, et al. Association between cell-derived microparticles and adverse events in patients with nonpulsatile left ventricular assist devices. J Heart Lung Transplant 2014;33:470-7. 20. Garcia BA, Smalley DM, Cho H, Shabanowitz J, Ley K, Hunt DF. The platelet microparticle proteome. J Proteome Res 2005;4:1516-21. 21. Smalley DM, Root KE, Cho H, Ross MM, Ley K. Proteomic discovery of 21 proteins expressed in human plasma-derived but not plateletderived microparticles. Thromb Haemost 2007;97:67-80. 22. Banfi C, Brioschi M, Wait R, et al. Proteome of endothelial cellderived procoagulant microparticles. Proteomics 2005;5:4443-55. 23. Hugel B, Martinez MC, Kunzelmann C, Freyssinet JM. Membrane microparticles: two sides of the coin. Physiology (Bethesda) 2005;20:22-7. 24. Morel O, Toti F, Hugel B, Freyssinet JM. Cellular microparticles: a disseminated storage pool of bioactive vascular effectors. Curr Opin Hematol 2004;11:156-64. 25. Horn P, Cortese-Krott MM, Amabile N, et al. Circulating microparticles carry a functional endothelial nitric oxide synthase that is decreased in patients with endothelial dysfunction. J Am Heart Assoc 2013;2:e003764. 26. Liu C, Zhao W, Christ GJ, Gladwin MT, Kim-Shapiro DB. Nitric oxide scavenging by red cell microparticles. Free Radic Biol Med 2013;65: 1164-73. 27. Madden JL, Drakos SG, Stehlik J, et al. Baseline red blood cell osmotic fragility does not predict the degree of post-LVAD hemolysis. ASAIO J 2014;60:524-8. 28. Costanzo MR, Dipchand A, Starling R, et al. The International Society of Heart and Lung Transplantation Guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2010;29:914-56. 29. Heiss C, Jahn S, Taylor M, et al. Improvement of endothelial function with dietary flavanols is associated with mobilization of circulating angiogenic cells in patients with coronary artery disease. J Am Coll Cardiol 2010;56:218-24. 30. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33:1787-847. 31. Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiographyʼs Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18: 1440-63. 32. Meyer C, Heiss C, Drexhage C, et al. Hemodialysis-induced release of hemoglobin limits nitric oxide bioavailability and impairs vascular function. J Am Coll Cardiol 2010;55:454-9. 33. Edvinsson ML, Uddman E, Andersson SE. Deteriorated function of cutaneous microcirculation in chronic congestive heart failure. Journal of geriatric cardiology : JGC 2011;8:82-7. 34. Andersson SE, Edvinsson ML, Edvinsson L. Cutaneous vascular reactivity is reduced in aging and in heart failure: association with inflammation. Clin Sci (Lond) 2003;105:699-707. 35. Bank AJ, Lee PC, Kubo SH. Endothelial dysfunction in patients with heart failure: relationship to disease severity. J Card Fail 2000;6:29-36. 36. Hasin T, Topilsky Y, Schirger JA, et al. Changes in renal function after implantation of continuous-flow left ventricular assist devices. J Am Coll Cardiol 2012;59:26-36.

12

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

37. Keymel S, Sichwardt J, Balzer J, et al. Characterization of the noninvasive assessment of the cutaneous microcirculation by laser Doppler perfusion scanner. Microcirculation 2010;17:358-66. 38. Hofmann U, Frantz S. How can we cure a heart “in flame”? A translational view on inflammation in heart failure. Basic Res Cardiol 2013;108:356. 39. Meyer B, Mortl D, Strecker K, et al. Flow-mediated vasodilation predicts outcome in patients with chronic heart failure: comparison with B-type natriuretic peptide. J Am Coll Cardiol 2005;46:1011-8. 40. Hambrecht R, Fiehn E, Weigl C, et al. Regular physical exercise corrects endothelial dysfunction and improves exercise capacity in patients with chronic heart failure. Circulation 1998;98:2709-15. 41. Cortese-Krott MM, Rodriguez-Mateos A, Sansone R, et al. Human red blood cells at work: identification and visualization of erythrocytic eNOS activity in health and disease. Blood 2012;120:4229-37. 42. Heiss C, Lauer T, Dejam A, et al. Plasma nitroso compounds are decreased in patients with endothelial dysfunction. J Am Coll Cardiol 2006;47:573-9. 43. Liu XM, Peyton KJ, Durante W. Physiological cyclic strain promotes endothelial cell survival via the induction of heme oxygenase-1. Am J Physiol Heart Circ Physiol 2013;304:H1634-43. 44. Blackman BR, Garcia-Cardena G, Gimbrone MA Jr. A new in vitro model to evaluate differential responses of endothelial cells to simulated arterial shear stress waveforms. J Biomech Eng 2002;124: 397-407. 45. Amir O, Radovancevic B, Delgado RM 3rd, et al. Peripheral vascular reactivity in patients with pulsatile vs axial flow left ventricular assist device support. J Heart Lung Transplant 2006;25:391-4. 46. Shechter M, Matetzky S, Feinberg MS, Chouraqui P, Rotstein Z, Hod H. External counterpulsation therapy improves endothelial function in

47.

48.

49.

50.

51.

52. 53.

54. 55.

56.

patients with refractory angina pectoris. J Am Coll Cardiol 2003;42: 2090-5. Moazami N, Fukamachi K, Kobayashi M, et al. Axial and centrifugal continuous-flow rotary pumps: a translation from pump mechanics to clinical practice. J Heart Lung Transplant 2013;32:1-11. Wang X, Tanus-Santos JE, Reiter CD, et al. Biological activity of nitric oxide in the plasmatic compartment. Proc Natl Acad Sci U S A 2004; 101:11477-82. Cortese-Krott MM, Kelm M. Endothelial nitric oxide synthase in red blood cells: key to a new erythrocrine function? Redox Biol 2014;2: 251-8. Vaughn MW, Huang KT, Kuo L, Liao JC. Erythrocyte consumption of nitric oxide: competition experiment and model analysis. Nitric Oxide 2001;5:18-31. Azarov I, Huang KT, Basu S, Gladwin MT, Hogg N, Kim-Shapiro DB. Nitric oxide scavenging by red blood cells as a function of hematocrit and oxygenation. J Biol Chem 2005;280:39024-32. Vaughn MW, Kuo L, Liao JC. Effective diffusion distance of nitric oxide in the microcirculation. Am J Physiol 1998;274:H1705-14. Kathiresan S, Gona P, Larson MG, et al. Cross-sectional relations of multiple biomarkers from distinct biological pathways to brachial artery endothelial function. Circulation 2006;113:938-45. Kuhn M. Endothelial actions of atrial and B-type natriuretic peptides. Br J Pharmacol 2012;166:522-31. Hoare GS, Birks EJ, Bowles C, Marczin N, Yacoub MH. In vitro endothelial cell activation and inflammatory responses in end-stage heart failure. J Appl Physiol (1985) 2006;101:1466-73. John R, Panch S, Hrabe J, et al. Activation of endothelial and coagulation systems in left ventricular assist device recipients. Ann Thorac Surg 2009;88:1171-9.

Macrovascular and microvascular function after implantation of left ventricular assist devices in end-stage heart failure: Role of microparticles.

The hemodynamic vascular consequences of implanting left ventricular assist devices (LVADs) have not been studied in detail. We investigated the effec...
3MB Sizes 0 Downloads 7 Views