Author + information
- Received March 22, 2016
- Revision received May 20, 2016
- Accepted June 20, 2016
- Published online October 10, 2016.
- Can Öztürk, MDa,
- Robert Schueler, MDa,
- Marcel Weber, MDa,
- Armin Welz, MDb,
- Nikos Werner, MDa,
- Georg Nickenig, MDa and
- Christoph Hammerstingl, MDa,∗ ()
- aDepartment of Medicine II, Heart Center Bonn, University Hospital Bonn, Bonn, Germany
- bDepartment of Cardiovascular Surgery, Heart Centre Bonn, University Hospital Bonn, Bonn, Germany
- ↵∗Reprint requests and correspondence:
Dr. Christoph Hammerstingl, Medizinische Klinik und Poliklinik II, Universitätsklinikum Bonn, Sigmund-Freud-Str. 25, 53105 Bonn, Germany.
Objectives This study shows the impact of secondary mitral regurgitation (sMR) and transcatheter mitral valve repair (TMVR) with the MitraClip system on sympathetic nerve activity (SNA).
Background An increase in SNA is associated with worse outcomes and limited survival in patients with chronic heart failure (CHF).
Methods Twenty CHF-patients without relevant sMR and 30 CHF patients with symptomatic sMR were enrolled prospectively. All patients underwent standardized laboratory testing and microneurography. Sixteen patients from the sMR group underwent the MitraClip procedure; 10 patients after TMVR and 9 untreated sMR patients completed 6 months of follow-up.
Results Comparing groups according to presence of sMR, we found no differences in left ventricular dimensions, and serum levels of N-terminal pro-brain natriuretic peptide (NT-proBNP) and noradrenaline; sMR was associated with increased MSNA (106 ± 60 burst/min vs. 74 ± 48 burst/min, d = 0.58), an impaired sympathetic baroreflex gain (10 ± 7 burst/mm Hg vs. 5 ± 5 burst/mm Hg, d = 0.61), and a higher heart rate (90 ± 27/beats/min vs. 78 ± 12/beats/min, d = 0.58). TMVR led to improved New York Heart Association functional class (d > 0.05), reduced levels of NT-proBNP (5,251 ± 3,760 pg/ml vs. 3,710 ± 2,464 pg/ml; d = 0.58) improvement in 6-minute walk test (204 ± 33 m vs. 288 ± 45 m, d = 0.64), but unchanged levels of noradrenaline. TMVR decreased MSNA burst-frequency (130 ± 78 bursts/min vs. 74 ± 21 bursts/min; d = 0.58) and baroreflex gain (7 ± 4 burst/mm Hg vs. 4 ± 1 burst/mm Hg; d = 0.61).
Conclusions In patients with CHF, concomitant sMR is associated with increased sympathetic nerve activity, which was independent from measured levels of NT-proBNP, noradrenaline, and left ventricular dimensions. Reduction of sMR with the MitraClip procedure reduced SNA and improved baroreflex gain, in line with improvements of functional capacity.
In patients with severe chronic heart failure (CHF) low cardiac output triggers an increase in sympathetic nerve activity (SNA) to maintain sufficient circulation. Increased SNA, on the other hand, has been shown to be associated with worse prognosis in CHF patients (1,2). SNA stimulates the production of renin, which leads to increased natrium retention from the renal tubuli. It induces systemic vasoconstriction leading to arterial hypertension and impaired peripheral perfusion. This effect potentially promotes hydropic decompensation, pulmonary congestion, and progressive left ventricular (LV) dilation. SNA can be determined by measurement of muscle SNA (MSNA), and baroreflex gain. Both parameters have been proven to be independent markers for adverse outcomes in different patient populations with CHF (3–5). An activated SNA has been shown of prognostic relevance in patients and animals with isolated primary mitral regurgitation (MR) (6,7).
Secondary MR (sMR) is a common finding in CHF patients with ischemic or nonischemic cardiomyopathy. The presence of sMR in heart failure patients is associated with poor prognosis (8) and the impact of surgical treatment of MR on SNA has been shown previously (9,10). It is unclear whether interventional reduction of sMR impacts on patient's prognosis and the effects of MitraClip procedure on SNA are unknown. Because SNA is an established independent risk marker for adverse outcome in CHF patients, we hypothesized that the presence of MR in CHF patients deteriorates SNA in CHF patients and that interventional reduction of MR might lead to amelioration of SNA. The objectives of this study were to: 1) compare SNA in CHF patients with and without sMR; and 2) examine the impact of transcatheter mitral valve repair (TMVR) on SNA.
Thirty consecutive patients presenting with symptomatic moderate to severe MR (MR grade >II) and 20 controls with CHF due to LV systolic dysfunction (left ventricular ejection fraction [LVEF] <40%) were prospectively enrolled to the study. Patients with CHF and without relevant MR were matched to sMR patients concerning baseline characteristics (age, sex, and body mass index), functional capacity, and cardiovascular comorbidities. All patients underwent standardized transthoracic echocardiography, clinical examination, 6-min walk test, routine laboratory testing, and microneurography. N-terminal pro-brain natriuretic peptide (NT-proBNP) and noradrenaline levels were determined from blood (EDTA or serum) in each patient. The 6-month follow-up after MitraClip included clinical examination, routine laboratory testing, transthoracic echocardiography, 6-min walk test, and repeated microneurography measurement.
The study was approved by the ethics committee of the University of Bonn and in concordance with the Declaration of Helsinki. All patients had to provide written informed consent before study inclusion.
Assessment of SNA
SNA was determined by measuring MSNA, baroreflex gain and neurohormonal activity. MSNA was assessed with microneurography (ADInstruments, Neuro Amp EX, Oxford, United Kingdom). After local disinfection a tungsten needle (200 μm) was introduced into the nervus peroneus longus close to the caput fibuli to record multiunit postganglionic sympathetic activity (Figure 1) (11). The recording started after 5 min resting in supine position to ensure standardization to the setting (Figure 1). Sympathetic muscle activity is defined by burst frequency (burst/min) and burst incidence (burst/100 beats) (12). Neural activity was amplified (50,000 to 100,000), band-passed (300 to 3,000 Hz) and integrated, as described previously (1,11).
During MSNA recording blood pressure was measured non-invasively by use of SOMNOtouch NIBP (SomnoMedics, Randersacker, Germany) consisting of a 4-channel electrocardiography and pulse meter, which was attached to the index finger of the patients and enables continuous blood pressure monitoring. Results from continuous blood pressure measurements were used for the calculation of baroreflex gain.
Cardiovascular reflexes, such as the baroreceptor heart rate reflex (baroreflex), regulate the hemodynamic response to parasympathetic and sympathetic signals. The baroreflex gain plays an important role in the development und progress of cardiovascular diseases and can be used for risk stratification after myocardial infarction, heart failure, or arrhythmias (13,14). Baroreflex gain is defined as the slope of the XY graphic curve from burst incidence and continuous diastolic blood pressure. For the determination of noradrenaline levels EDTA blood was centrifuged at 3000 RPM for 10 min; the supernatant was removed and deep frozen (-79°C). NT-proBNP was determined from lithium heparin plasma.
Echocardiographic assessment of LV function was done following current recommendations and guidelines (15). Parasternal long and short axis and apical 4- and 2-chamber, as well as apical long axis, views were transthoracic recorded (transthoracic echocardiogram: transthoracic echocardiography). The LVEF was calculated by Simpsons rule from 4- and 2-chamber views. Assessment of MR consists of determination of proximal isovelocity surface area, effective regurgitant orifice area, as well as vena contracta width and regurgitant volume, where applicable, to specify severity of MR. MR was graded according to the recommendations of the European Society of Cardiology as “mild,” “moderate,” “moderate to severe,” or “severe” (16,17). Transthoracic echocardiography were performed with a commercially available echocardiographic system (iE 33, Philips Medical Systems, Andover, Massachusetts).
sMR was defined as mostly central MR with global (symmetric) or regional (eccentric) LV dilation and reduced LVEF despite structurally normal MV leaflets.
Exploratory data analysis was performed and no adjustment was made for multiple tests. Normal distribution of continuous variables was examined using the Kolmogorov–Smirnov test. Continuous data were expressed as mean ± SD. Because of the small patient number and to prevent inflating type I error rate we did not calculate p-values from any kind of test. Calculated standardized differences were calculated as the mean divided by the standard deviation of a difference between 2 values from the 2 groups. Following Cohen (18), 3 effect size indices (0.2, 0.5, 0.8) were chosen to represent small, medium, and large effect sizes. According to current evidence, large and medium effects (≥0.5) can be assessed with the “naked eye,” a small effect (≤0.2) is more difficult to estimate and obviously not of clinical relevance (18,19). The method of Bland and Altman was used for the assessment of interobserver agreement. For the assessment of intraobserver variability, 20 randomly chosen patients were analyzed by the same investigator twice. Intraobserver variability was evaluated by intraclass correlation coefficient for total agreement, with good agreement being defined as >0.80. Mean values and standard deviations between the measurements were obtained and total agreement among the observation was calculated using intraclass correlation analysis.
From January 2014 to July 2015, 50 consecutive CHF patients were enrolled, including 20 patients (71.3 ± 8.5 years; 85% male) without relevant MR (CHF group) and 30 patients (79.2 ± 7.4 years; 73.3% male) with CHF and symptomatic sMR (sMR group). Sixteen patients underwent successful TMVR with the MitraClip system. Six patients presented with unsuitable anatomy for MitraClip procedure (insufficient coaptation length [n = 3], leaflet calcification [n = 1], inability to perform transesophageal echocardiography due to esophageal stenosis [n = 1], and insufficient transesophageal echocardiography image quality [n = 1]). In 4 patients, the heart team decision voted for surgical MV treatment. The remaining 4 patients declined either surgical or interventional treatment. During follow-up, 2 of the patients treated with MitraClip died and 4 patients declined further study participation. Finally, 10 patients with implanted MitraClip and 9 untreated sMR patients completed 6 months of follow-up procedures, including microneurography measurement (Figure 2).
The CHF and sMR patients did not differ relevantly concerning demographic baseline characteristics, blood pressure, medication, and New York Heart Association functional class (Table 1).
As determined by echocardiography, the LVEF was 37.6 ± 15.8% in the overall cohort. Patients with sMR presented with better LV systolic function as compared with the CHF group (sMR: 43 ± 18%; CHF: 29 ± 7%; d = 0.64) (Table 2).
Patients with sMR underwent treatment for moderate-severe MR in 93.3% and severe MR in 10% of cases (effective regurgitant orifice area: 0.3 ± 0.1 cm; vena contracta width: 0.7 ± 0.1 cm; regurgitant volume: 53 ± 19 ml/beat) (Table 2).
MSNA in patients with and without sMR
When compared with normal values, we found increased MSNA (93 ± 57 burst/min [normal values: 20 ± 3]) and impaired baroreflex gain (7 ± 4 burst/mm Hg [normal values: 2 ± 0.2]), the average heart rate was 85 beats/min.
The presence of sMR alone was associated with increased MSNA (sMR: 106 ± 59 burst/min; CHF: 74 ± 48.1 burst/min; d = 0.58), an impaired sympathetic baroreflex gain (sMR: 10 ± 7 burst/mm Hg; CHF: 5 ± 5 burst/mm Hg; d = 0.61), and a higher heart rate (sMR: 90 ± 27/min; CHF: 78 ± 12/min; d = 0.58) (Table 3). The MSNA burst incidence was increased in sMR patients but not different to the CHF group (sMR: 119 ± 63; CHF: 93 ± 58 burst/100 beats; d = 0.48).
Levels of NT-proBNP (88,323 ± 13,907 pg/ml) and noradrenaline serum levels (883 ± 493 ng/l) were increased in the overall cohort without differences between groups regarding levels of NT-proBNP (sMR: 6,577 ± 6,816 pg/dl; CHF: 13,010 ± 21,495 pg/dl; d = 0.37) and noradrenaline (sMR: 882 ± 531 ng/l; CHF: 885 ± 452 ng/l; d = 0.03). Increased MSNA was associated with an elevated NT-proBNP (r = 0.69; p = 0.03), but not with renal function (r = 0.03; p = 0.9) (Figures 3A and 3B).
Sympathetic activity and functional outcomes after 6 months of follow-up
Ten patients of the sMR group who underwent TMVR with the MitraClip and 9 untreated patients with sMR completed 6 months of follow-up procedures with complete assessment of SNA. Medical treatment remained unchanged during follow-up.
Echocardiography showed stable MR reduction in all patients after the MitraClip procedure (effective regurgitant orifice area: 0.3 ± 0.1 cm2 vs. 0.2 ± 0.1 cm2; d = 0.48; regurgitant volume: 53 ± 19 vs. 30 ± 8 ml/beat; d = 0.64) (Figure 4) and decrease in left atrial volumes (131 ± 63 ml vs. 78 ± 23 ml; d = 0.58); LVEF and LV volumes remained unchanged after 6 months.
Functional New York Heart Association class improved during follow-up (2.8 ± 0.4 vs. 1.6 ± 0.5; d = 0.61), as well as the 6-minute walking distances (204 ± 33 m vs. 288 ± 45 m; d = 0.64). There was no difference in systolic (TMVR: 129 ± 65 mm Hg; sMR: 139 ± 44 mm Hg; d = 0.07) or diastolic blood pressure (TMVR: 90 ± 33 mm Hg; sMR: 92 ± 48 mm Hg; d = 0.07) in patients with sMR, regardless of MitraClip treatment.
Concerning MSNA measurements, sMR patients that underwent TMVR showed decreased MSNA burst-frequency (130 ± 78 bursts/min vs. 74 ± 21 bursts/min; d = 0.58), baroreflex gain (7 ± 4 bursts/mm Hg vs. 4 ± 1 bursts/mm Hg; d = 0.64) and burst incidence (143 ± 88 bursts/100 beats vs. 113 ± 34 bursts/100 beats; d = 0.48). Patients with untreated sMR showed no changes in MSNA (d < 0.5) (Table 4).
NT-proBNP levels (6,577 ± 6,816 pg/ml vs. 3,710 ± 2,464 pg/ml; d = 0.58) and noradrenaline levels were reduced in patients 6 months after MitraClip treatment (882 ± 530 ng/l vs. 624 ± 351 ng/l; d = 0.30). In untreated patients, we found no relevant changes regarding NT-proBNP (6,577 ± 6,816 vs. 6,784 ± 2,789; d = 0.03) or noradrenalin (882 ± 530 vs. 746 ± 99; d = 0.17) serum levels.
The main findings of our study were that concomitant sMR in CHF patients promotes impaired SNA as determined by microneurography, whereas laboratory measures on neurohumoral activity were not altered relevantly in the presence of sMR. After TMVR, SNA improved to comparable levels of patients with CHF but without relevant MR. Because impaired SNA has been shown to be associated with worse prognosis in CHF patients, these findings may help to explain the impact of sMR on survival in typical CHF patients.
Impact of sympathetic activity on prognosis
MSNA is a known independent marker of adverse clinical outcome in patients with CHF, obstructive sleep apnea syndrome, arterial hypertension, and chronic anemia (20–22). Mehta et al. (6) found an activation of SNA in patients with MR compared with healthy controls as a compensatory mechanism for chronic LV volume overload. However, changes in SNA after surgical repair were not homogenous and related to systolic LV function and dimensions before surgery, but not procedural success. Because all patients in our study had impaired LV function, the results are not comparable to the study by Mehta et al. (6). Furthermore, microneurography was not used in the studies by Mehta et al. for the determination of SNA (9).
Elevated SNA induces tachycardia and arterial hypertension due to amplified stimulation of adrenergic receptors. Therefore, increased SNA potentially decreases myocardial tissue perfusion, increases the heart rate, and thus deteriorates systolic LV function. Tsutsui et al. (7) found a chronic β-blockade to have beneficial effects on LV function in dogs with chronic MR. Ahmed et al. (10) showed that β-blockers improve systolic function in patients with asymptomatic primary MR, indicating on activated SNA in primary MR patients and proving a negative influence of activated SNA on LV function in those patients. Because β-blocker therapy was not different in the patient groups included in our study, we are reluctant to compare our findings with the study by Ahmed et al. (10).
Of note, Barretto et al. (11) examined 22 patients with CHF and showed that impaired MSNA was an independent predictor for increased mortality. Furthermore, Leimbach et al. (2) demonstrated elevated SNA in 16 moderate to severe CHF patients in comparison with 19 healthy controls. They showed a significant correlation between plasma norepinephrine and SNA levels.
SNA after interventional valve treatment
Studies on SNA in patients with sMR are scarce. Hu et al. (23) found SNA to be related to myxomatous degeneration in mitral valve prolapse, which propagates the disease severity. In an early study, Ashino et al. (24) demonstrated decreased MSNA and normalized baroreflex gain in a subset of 10 patients with mitral stenosis 1 week after interventional mitral valvuloplasty. In concordance with our findings, this group found no difference in noradrenaline levels between patients with mitral stenosis and a healthy group (24).
More recently, Dumonteil et al. (25) examined 14 patients undergoing transcatheter aortic valve replacement and found normalized MSNA values 30 days after transcatheter aortic valve replacement. The impact of TMVR on SNA in sMR patients is not known.
Interventional edge-to-edge repair with the MitraClip system (Abbott Vascular, Abbott Park, Illinois) is a treatment option in surgical high-risk patients with symptomatic MR. MitraClip procedure improves heart failure symptoms in sMR and primary MR (26). The impact of MitraClip procedure on survival is discussed controversially and the underlying mechanisms for a beneficial influence on survival are unclear.
In our study, we found evidence that sMR is associated with impaired SNA in patients with CHF and TMVR by use of the MitraClip system decreased SNA in treated patients. These findings were independent of changes in LV dimensions and laboratory measures on neurohumoral activity.
Our study is limited by its single side character and limited patient number; the presented results, therefore, are preliminary and must be reconfirmed in independent patient cohorts. We, furthermore, completed follow-up only in a small number of patients. A longer follow-up time and greater patient numbers might result in different results. Differences in specific measures of SNA (i.e., levels of noradrenaline) were small, which could, as well, be due to the small sample size of the study groups.
In patients with CHF, sMR is associated with increased SNA, which was independent from measured levels of NT-proBNP and noradrenaline. Reduction of MR with the MitraClip system reduces SNA and improves baroreflex gain.
WHAT IS KNOWN? In patients with severe CHF, low cardiac output triggers an increase in SNA to maintain sufficient circulation.
WHAT IS NEW? Patients with sMR had higher sympathetic activity levels when compared with patients with CHF but without MR. SNA improved after interventional edge-to-edge repair.
WHAT IS NEXT? Determination of SNA might contribute to a better understanding of the effects of interventional treatment of sMR on patient's functional outcome and survival.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Drs. Öztürk and Schueler contributed equally to this work.
- Abbreviations and Acronyms
- chronic heart failure
- left ventricular
- left ventricular ejection fraction
- mitral regurgitation
- muscle sympathetic nerve activity
- N-terminal pro-brain natriuretic peptide
- secondary mitral regurgitation
- sympathetic nerve activity
- transcatheter mitral valve repair
- Received March 22, 2016.
- Revision received May 20, 2016.
- Accepted June 20, 2016.
- American College of Cardiology Foundation
- Tank J.,
- Diedrich A.,
- Szczech E.,
- et al.
- Leimbach W.N.,
- Wallin B.G.,
- Victor R.G.,
- et al.
- Goya T.T.,
- Silva R.F.,
- Guerra R.S.,
- et al.
- Groehs R.V.,
- Toschi-Dias E.,
- Antunes-Correa L.M.,
- et al.
- Mehta R.H.,
- Supiano M.A.,
- Grossman P.M.,
- et al.
- Ahmed M.I.,
- Aban I.,
- Lloyd S.,
- et al.
- Hamdan M.H.,
- Joglar J.A.,
- Page R.L.,
- et al.
- Lurz P.,
- Serpytis R.,
- Blazek S.,
- et al.
- Cohen J.
- (2008) Interpreting estimates of treatment effects: implications for managed care. PT 33:700–711.
- Franchitto N.,
- Despas F.,
- Labrunee M.,
- et al.
- Hamaoka T.,
- Murai H.,
- Okabe Y.,
- et al.
- Shimizu T.,
- Takahashi Y.,
- Kogawa S.,
- et al.
- Ashino K.,
- Gotoh E.,
- Sumita S.,
- et al.
- Dumonteil N.,
- Vaccaro A.,
- Despas F.,
- et al.
- Feldman T.,
- Kar S.,
- Elmariah S.,
- et al.