Abstract
Background
Risk prediction in patients with severe aortic stenosis (AS) undergoing transcatheter aortic valve implantation (TAVI) is challenging. Development of novel markers for patient risk assessment is of great clinical value. Deceleration capacity (DC) of heart rate is a strong risk predictor in post‐infarction patients.
Hypothesis
DC provides prognostic information in patients undergoing TAVI.
Methods
We enrolled 374 consecutive patients with severe AS undergoing TAVI. All patients received 24‐hour Holter recording or continuous heart‐rate monitoring to assess DC before intervention. Primary endpoint was all‐cause mortality after 1 year.
Results
Forty‐nine patients (13.1%) died within 1 year. DC was significantly lower in nonsurvivors than in survivors (1.2 ± 4.8 ms vs 3.3 ± 2.9 ms; P < 0.001), whereas the logistic EuroSCORE and EuroSCORE II were comparable between groups (logistic EuroSCORE: 27.3% ± 17.0% vs 22.9% ± 14.2%; P = 0.122; EuroSCORE II: 8.0% ± 6.9% vs 6.7% ± 4.8%, P = 0.673). One‐year mortality in the 116 patients with impaired DC (<2.5 ms) was significantly higher than in patients with normal DC (23.3% vs 8.5%; P < 0.001). In multivariate Cox regression analysis that included DC, sex, paroxysmal atrial fibrillation, hemoglobin level before TAVI, and logistic EuroSCORE, DC was the strongest predictor of 1‐year mortality (hazard ratio: 0.88, 95% confidence interval: 0.85‐0.94, P < 0.001). DC yielded an AUC in the ROC analysis of 0.645.
Conclusions
DC of heart rate is a strong and independent predictor of 1‐year mortality in patients with severe AS undergoing TAVI.
Keywords: TAVI, Risk Markers, Deceleration Capacity, EuroSCORE, Mortality, Cardiac Autonomic Dysfunction
1. INTRODUCTION
Aortic valve stenosis (AS) is the most common valvular heart disease in the industrialized countries.1, 2 Its treatment has been revolutionized by the introduction of transcatheter aortic valve implantation (TAVI).3
The decision for adequate therapy (conservative treatment vs TAVI vs conventional aortic valve replacement)4, 5 is based on established score systems such as the logistic EuroSCORE (log ES) and EuroSCORE II (ES II).6, 7 It is usually recommended that older patients with higher scores undergo TAVI or even conservative treatment. However, the 1‐year mortality rate after TAVI was as high as 24.7% in the Placement of Aortic Transcatheter Valve (PARTNER) trial, and some patients may be too sick to benefit from the intervention.8 Therefore, this patient cohort needs careful evaluation of the risk‐to‐benefit ratio of intervention.9 Conventional risk scores as mentioned above are validated to predict perioperative mortality in low‐ or intermediate‐risk patients undergoing surgical treatment, but they show moderate predictive value in patients undergoing TAVI.10 Therefore, the identification of novel markers that allow for risk stratification of patients with severe AS undergoing TAVI is of great general interest.
The assessment of cardiac autonomic dysfunction provides excellent information about the regulatory capabilities of the cardiovascular system.11 The strong and independent prognostic value of different markers of cardiac autonomic dysfunction has been already demonstrated in patients with heart failure or myocardial infarction (MI).12, 13 Deceleration capacity (DC) of heart rate, as one of these autonomic parameters, is representative of tonic vagal activity. DC quantifies the mean amplitude of all deceleration‐related oscillations of heart rate observed in the recording period.11, 14, 15
The aim of this study was to test whether DC provides prognostic information in patients with severe AS undergoing TAVI.
2. METHODS
2.1. Enrollment
We studied consecutive symptomatic patients with severe AS between December 2010 and January 2014 who underwent TAVI at our tertiary university center in Tübingen, Germany. Severe AS was defined as either presence of an aortic valve area (AVA) <1 cm2, a mean aortic gradient >40 mm Hg, or jet velocity >4.0 m/s, evaluated both invasively and by echocardiography. The decision for TAVI was made by an interdisciplinary heart team, who were not involved in the study, based on the current guidelines.4, 5, 6 The patient's electrocardiogram (ECG) was recorded either by 24‐hour Holter recordings (CardioMem CM 3000; Getemed, Teltow, Germany) or by monitoring devices (DASH 4000/5000 or Teleguard; GE, Fairfield, CT) between the day of hospital admission and the intervention. All patients presenting in sinus rhythm during recording were included. Patients with known paroxysmal atrial fibrillation (AF) were included if they had a sufficient period of sinus rhythm for calculation of DC (see Supporting Information, Figure 1S, in the online version of this article).
2.2. Assessment of DC
Technical details of the automated assessment of DC have been described elsewhere.16 The ECG recordings were checked for AF using a validated automated algorithm.17 Sections of AF were excluded.
Recordings exhibiting permanent AF and noisy, low‐quality signals were excluded from further analysis (see Supporting Information, Figure 1S, in the online version of this article).
Assessment of DC was performed by applying a signal‐processing algorithm called phase‐rectified signal averaging, which is capable of extracting periodic components out of nonstationary, noisy signals.18 Briefly, DC calculation is performed in 5 steps. First, R‐R intervals are identified that are longer than their predecessors; these are defined as anchors. Second, intervals surrounding the anchors, which may overlap, are defined. In the third and fourth steps, segments are aligned at the anchors and subsequently averaged. Fifth, the phase‐rectified signal averaging signal is quantified by Haar wavelet analysis. Based on previous investigations, patients were stratified according to DC into high‐risk (DC <2.5 ms) and low‐risk (DC ≥2.5 ms) categories.14
2.3. Conventional risk predictors
Left ventricular ejection fraction (LVEF), pulmonary artery pressure, and peak and mean aortic gradients were assessed both invasively and by echocardiography. History of paroxysmal AF was evaluated in every patient.
According to log ES criteria, age, sex, chronic pulmonary disease, chronic renal failure, extracardiac arteriopathy, neurological dysfunction, previous cardiac surgery, active endocarditis, critical perioperative state, unstable angina, LVEF, recent MI, pulmonary hypertension, surgery different from isolated coronary artery bypass grafting, surgery on the thoracic aorta, post‐infarct septal rupture, and emergency status were defined.7
To calculate the ES II, age, sex, chronic renal failure, extracardiac arteriopathy, poor mobility, previous cardiac surgery, chronic lung disease, active endocarditis, critical perioperative state, insulin‐dependent diabetes mellitus, New York Heart Association class, Canadian Cardiovascular Society class, LVEF, recent MI, pulmonary hypertension, urgency, weight of the intervention, and surgery on the thoracic aorta were assessed.19
Furthermore, serum creatinine (Cr) and hemoglobin (Hgb) levels were assessed in every patient at baseline.
2.4. Study endpoints
The primary endpoint was all‐cause mortality 1 year after TAVI. The secondary endpoint was all‐cause mortality after 30 days.
2.5. Follow‐up
Intrahospital deaths were recorded by the electronic information system. Patients were followed up 1 year after TAVI in our outpatient clinic or by telephone contact.
2.6. Statistical analysis
Continuous variables are presented as mean ± SD and were compared using the Mann‐Whitney U test. Qualitative data are presented both as absolute value and as percentages and were analyzed using the χ2 test. Receiver operator characteristic (ROC) curves were constructed for DC by plotting specificity vs sensitivity. ROC curves were quantified by the area under the curve (AUC). The association of risk variables with the primary endpoint was calculated by univariate and multivariate Cox regression analyses. Both parameters with significant difference in the baseline analysis and conventional risk factors were included. Mortality rates were estimated by the Kaplan‐Meier method.20 Hazard ratios (HRs) are presented with 95% confidence intervals (CIs). Differences were regarded as statistically significant if the P value was <0.05. Statistical analyses were performed using SPSS version 23.0 (IBM Corp., Armonk, NY) and CRAN R 3.3.0 (R Foundation for Statistical Computing, http://www.r-project.org).
3. RESULTS
Between December 2010 and January 2014, a total of 422 patients underwent TAVI. Due to either absence of sinus rhythm or noisy, low‐quality ECG signals, 48 patients were excluded, leaving 374 patients enrolled in the study (see Supporting Information, Figure 1S, in the online version of this article). Clinical and demographic characteristics of the study cohort are shown in Table 1. Mean age was 82.3 ± 6.5 years; 50.3% were female. Insulin‐dependent diabetes mellitus affected 11.8% of the patients, and 245 patients (65.5%) revealed a history of coronary artery disease. Mean AVA was 0.68 ± 0.17 cm2 and mean aortic gradient was 43.0 ± 15.7 mm Hg. The CoreValve system (Medtronic, Minneapolis, MN) was implanted in 294 patients, and 80 patients were treated with Sapien or Sapien XT valves (Edwards Lifesciences, Irvine, CA). The mean log ES was 23.5% ± 14.6%, the mean ES II was 7.8% ± 6.7%, and 40.4% of the study population had paroxysmal AF.
Table 1.
Characteristics of the study population with regard to DC status
| All Patients, N = 374 | DC <2.5 ms, n = 116 | DC ≥2.5 ms, n = 258 | P Value | |
|---|---|---|---|---|
| Age, y | 82.3 ± 6.5 | 82.8 ± 5.8 | 82.1 ± 6.8 | 0.624 |
| Female sex | 188 (50.3) | 46 (39.7) | 142 (55.0) | 0.006 |
| Arterial HTN | 305 (81.6) | 96 (82.8) | 209 (81) | 0.686 |
| Extracardiac arteriopathy | 78 (20.9) | 30 (25.9) | 48 (18.6) | 0.110 |
| CAD | 245 (65.5) | 81 (69.8) | 164 (63.6) | 0.239 |
| Insulin‐dependent DM | 44 (11.8) | 18 (15.5) | 26 (10.1) | 0.131 |
| Paroxysmal AF | 151 (40.4) | 58 (50) | 93 (36) | 0.012 |
| Poor mobility | 25 (6.7) | 8 (6.9) | 17 (6.6) | 0.912 |
| Urgency | 27 (7.2) | 5 (4.3) | 22 (8.5) | 0.145 |
| Previous MI | 63 (16.8) | 24 (20.7) | 39 (15.1) | 0.183 |
| Previous cardiac surgery | 171 (45.7) | 59 (50.9) | 112 (43.4) | 0.181 |
| Chronic pulmonary disease | 44 (11.8) | 18 (15.5) | 26 (10.1) | 0.131 |
| Active endocarditis | 1 (0.3) | 0 (0.0) | 1 (0.4) | 0.502 |
| Critical preoperative state | 3 (0.8) | 0 (0.0) | 3 (1.2) | 0.244 |
| Laboratory parameters | ||||
| GFR, mL/min | 66.3 ± 27.65 | 64.64 ± 27.2 | 67.0 ± 27.86 | 0.292 |
| Serum Cr, mg/dL | 1.27 ± 1.02 | 1.29 ± 0.86 | 1.26 ± 1.09 | 0.068 |
| Hgb, g/dL | 12.3 ± 1.58 | 12.5 ± 1.70 | 12.35 ± 0.09 | 0.594 |
| Symptoms | ||||
| NYHA class ≥ II | 277 (74.1) | 92 (79.3) | 185 (71.7) | 0.121 |
| CCS class IV | 5 (1.2) | 1 (0.9) | 4 (1.6) | 0.592 |
| Syncope | 51 (13.6) | 12 (10.3) | 39 (15.1) | 0.214 |
| Type of TAVI | ||||
| CoreValve | 294 (78.6) | 100 (86.2) | 194 (75.2) | 0.016 |
| Edwards Sapien/Sapien XT | 80 (21.4) | 16 (13.8) | 64 (24.8) | 0.016 |
| Hemodynamic parameters | ||||
| LVEF, % | 52.0 ± 11.5 | 50.4 ± 12.0 | 52.7 ± 11.2 | 0.035 |
| AVA, cm2 | 0.68 ± 0.17 | 0.66 ± 0.18 | 0.69 ± 0.17 | 0.181 |
| Pmax, mm Hg | 71.2 ± 25.3 | 70.24 ± 25.8 | 71.7 ± 25.1 | 0.608 |
| Pmean, mm Hg | 43.0 ± 15.7 | 42.32 ± 15.9 | 43.27 ± 15.6 | 0.583 |
| PaPSys, mm Hg | 34.16 ± 12.94 | 34.9 ± 13.2 | 33.8 ± 12.6 | 0.366 |
| Conventional risk variables | ||||
| Logistic EuroSCORE, % | 23.5 ± 14.6 | 25.63 ± 14.6 | 22.5 ± 14.6 | 0.025 |
| EuroSCORE II, % | 7.8 ± 6.7 | 8.32 ± 6.6 | 7.55 ± 6.64 | 0.191 |
| Endpoints | ||||
| Nonsurvivors after 1 year | 49 (13.1) | 27 (23.3) | 22 (8.5) | <0.001 |
| Nonsurvivors after 30 days | 13 (3.5) | 7 (6.0) | 6 (2.3) | 0.070 |
Abbreviations: AF, atrial fibrillation; AVA, aortic valve area; CAD, coronary artery disease; CCS, Canadian Cardiovascular Society; COPD, chronic obstructive pulmonary disease; Cr, creatinine; DC, deceleration capacity of heart rate; DM, diabetes mellitus; GFR, glomerular filtration rate; Hgb, hemoglobin; HTN, hypertension; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association; PaPSys, pulmonary artery systolic pressure; Pmax, maximal aortic valve gradient; Pmean, mean aortic valve gradient; sCr, serum creatinine; SD, standard deviation; TAVI, transcatheter aortic valve implantation.
Data are presented as n (%) or mean ± SD.
All‐cause mortality 1 year after TAVI was 13.1%. Of the 374 patients, 116 had an impaired DC (<2.5 ms). The 1‐year mortality rate in patients with impaired DC was significantly higher than in patients with a DC ≥2.5 ms (23.3% vs 8.5%, P < 0.001; Table 1, Figure 1). Mortality after 30 days marginally missed significant difference between the 2 groups (6.0% vs 2.3%; P = 0.070). Besides sex (39.7% vs 55.0%; P = 0.006) and LVEF (50.4% vs 52.7%; P = 0.035), components of the ES II were comparable.
Figure 1.
Kaplan‐Meier survival curves. Cumulative 1‐year mortality of patients undergoing TAVI stratified by DC. Abbreviations: CI, confidence interval; DC, deceleration capacity of heart rate; TAVI, transcatheter aortic valve implantation.
DC was significantly lower in the nonsurvivor group as compared with survivors (1.2 ± 4.8 ms vs 3.3 ± 2.9 ms, respectively, P < 0.001; Table 2). Conventional risk factors such as the log ES (22.9% ± 14.2% vs 27.3% ± 17.0%; P = 0.122), the ES II (6.7% ± 4.8% vs 8.0% ± 6.9%; P = 0.673), LVEF (51.7% ± 12.9% vs 52.0% ± 11.3%; P = 0.999) or pulmonary artery systolic pressure (36.5 ± 13.6 mm Hg vs 33.9 ± 12.9 mm Hg; P = 0.068) were comparable between nonsurvivors and survivors. Comparison of serum Cr and Hgb pre‐intervention revealed significant difference between nonsurvivors and patients who survived beyond 1 year after TAVI (Cr, 1.5 ± 1.0 mg/dL vs 1.2 ± 1.0 mg/dL, P = 0.008; Hgb, 11.8 ± 1.8 g/dL vs 12.4 ± 1.6 g/dL, P = 0.005). Nonsurvivors had a higher prevalence of paroxysmal AF before TAVI than did survivors (61.2% vs 37.2%; P < 0.001).
Table 2.
Characteristics of survivors and nonsurvivors with regard to the primary and secondary endpoints
| 1 year after TAVI | Survivors, n = 325 | Nonsurvivors, n = 49 | P Value |
|---|---|---|---|
| Age, y | 82.3 ± 6.5 | 82.3 ± 6.4 | 0.94 |
| Female sex | 171 (52.6) | 17 (34.7) | 0.019 |
| LVEF, % | 52.0 ± 11.3 | 51.7 ± 12.9 | 0.999 |
| PaPSys, mm Hg | 33.9 ± 12.9 | 36.5 ± 13.6 | 0.068 |
| Serum Cr, mg/dL | 1.2 ± 1.0 | 1.5 ± 1.0 | 0.008 |
| Serum Hgb, g/dL | 12.4 ± 1.6 | 11.8 ± 1.8 | 0.005 |
| Chronic renal failure | 256 (78.8) | 38 (77.6) | 0.846 |
| Paroxysmal AF | 121 (37.2) | 30 (61.2) | <0.001 |
| Logistic EuroSCORE, % | 22.9 ± 14.2 | 27.3 ± 17.0 | 0.122 |
| EuroSCORE II, % | 8.0 ± 6.9 | 6.7 ± 4.8 | 0.673 |
| DC, ms | 3.3 ± 2.9 | 1.2 ± 4.8 | <0.001 |
| 30 days after TAVI | Survivors, n = 361 | Nonsurvivors, n = 13 | |
|---|---|---|---|
| Age, y | 82.3 ± 6.5 | 82.6 ± 6.3 | 0.86 |
| Female sex | 182 (50.4) | 6 (46.2) | 0.76 |
| LVEF, % | 52.1 ± 16.9 | 48.3 ± 11.3 | 0.664 |
| PaPSys, mm Hg | 34.1 ± 13.0 | 34.5 ± 9.9 | 0.669 |
| Serum Cr, mg/dL | 12.34 ± 1.6 | 11.7 ± 2.1 | 0.075 |
| Serum Hgb, g/dL | 1.3 ± 1.0 | 1.7 ± 1.1 | 0.053 |
| Chronic renal failure | 283 (78.4) | 11 (84.6) | 0.591 |
| Paroxysmal AF | 143 (39.6) | 8 (61.5) | 0.11 |
| Logistic EuroSCORE, % | 23.1 ± 14.3 | 33.3 ± 19.8 | 0.04 |
| EuroSCORE II, % | 7.8 ± 6.7 | 8.1 ± 5.8 | 0.516 |
| DC, ms | 3.1 ± 3.3 | 1.4 ± 3.0 | 0.068 |
Abbreviations: AF, atrial fibrillation; Cr, creatinine; DC, deceleration capacity of heart rate; Hgb, hemoglobin; LVEF, left ventricular ejection fraction; PaPSys, pulmonary artery systolic pressure.
Data are presented as n (%) or mean ± SD.
All‐cause mortality 30 days after TAVI was 3.5%. DC was lower in the nonsurvivor group but did not reach statistical significance (1.4 ± 3.0 ms vs 3.1 ± 3.3 ms, P = 0.068; Table 2). The ES II was comparable between nonsurvivors and survivors (8.1% ± 5.8% vs 7.8% ± 6.7%; P = 0.516). However, the log ES was significantly higher in the nonsurvivor group (33.3% ± 19.8% vs 23.1% ± 14.3%; P = 0.04).
In a univariate Cox regression analysis, sex (HR: 0.5, 95% CI: 0.28‐0.90, P = 0.022), paroxysmal AF (HR: 2.49, 95% CI: 1.40‐4.40, P = 0.002), serum Hgb level before TAVI (HR: 0.79, 95% CI: 0.67‐0.95, P = 0.01), the log ES (HR: 1.02, 95% CI: 1.00‐1.04, P = 0.036), and DC (HR: 0.88, 95% CI: 0.84‐0.93, P < 0.001) were significantly associated with 1‐year mortality (Table 3). In a multivariate Cox regression analysis, which included all parameters being statistically significant in the univariate calculation, DC was the strongest predictor for 1‐year‐mortality (HR: 0.88, 95% CI: 0.85‐0.94, P < 0.001) and independent from sex (HR: 0.49, 95% CI: 0.27‐0.88, P = 0.018), paroxysmal AF (HR: 2.09, 95% CI: 1.16‐3.8, P = 0.015), and Hgb level before TAVI (HR: 0.80, 95% CI: 0.68‐0.96, P = 0.014), which were also independent predictors of mortality. The log ES was not associated with 1‐year mortality (HR: 1.00, 95% CI: 0.99‐1.03, P = 0.371) and therefore no independent risk predictor in the multivariate analysis.
Table 3.
Univariate and multivariate Cox regression analysis for prediction of 1‐year mortality in patients undergoing TAVI
| Univariate Analysis | Multivariate Analysis | |||||
|---|---|---|---|---|---|---|
| HR (95% CI) | z | P Value | HR (95% CI) | z | P Value | |
| Variable | ||||||
| Female sex | 0.50 (0.28‐0.90) | 5.27 | 0.022 | 0.49 (0.27‐0.88) | 5.57 | 0.018 |
| Paroxysmal AF | 2.49 (1.40‐4.40) | 9.65 | 0.002 | 2.09 (1.16‐3.80) | 5.94 | 0.015 |
| Serum Cr | 1.15 (0.94‐1.40) | 1.85 | 0.174 | |||
| Serum Hgb | 0.79 (0.67‐0.95) | 6.71 | 0.01 | 0.80 (0.68‐0.96) | 6.06 | 0.014 |
| Chronic renal failure | 1.12 (0.57‐2.18) | 0.10 | 0.750 | |||
| Logistic EuroSCORE | 1.02 (1.00‐1.04) | 4.38 | 0.036 | 1.00 (0.99‐1.03) | 0.80 | 0.371 |
| EuroSCORE II | 0.97 (0.92‐1.02) | 1.36 | 0.243 | |||
| LVEF | 0.98 (0.97‐1.02) | 0.06 | 0.804 | |||
| PaPSys | 0.99 (0.97‐1.01) | 0.25 | 0.615 | |||
| DC | 0.88 (0.84‐0.93) | 22.75 | <0.001 | 0.88 (0.85‐0.94) | 18.2 | <0.001 |
Abbreviations: AF, atrial fibrillation; CI, confidence interval; Cr, creatinine; DC, deceleration capacity of heart rate; Hgb, hemoglobin; HR, hazard ratio; LVEF, left ventricular ejection fraction; PaPSys, pulmonary artery systolic pressure; TAVI, transcatheter aortic valve implantation.
Figure 2 shows the ROC curve for prediction of 1‐year mortality for DC. DC yielded an AUC of 0.645 (95% CI: 0.566‐0.740, P = 0.001), the log ES an AUC of 0.568 (95% CI: 0.478‐0.659, P = 0.122), and the ES II an AUC of 0.481 (95% CI: 0.403‐0.559, P = 0.673).
Figure 2.
ROC curve of DC for prediction of 1‐year‐mortality after TAVI. Abbreviations: CI, confidence interval; DC, deceleration capacity of heart rate; ROC, receiver operating characteristic; TAVI, transcatheter aortic valve implantation.
4. DISCUSSION
This study shows that the DC of heart rate provides important prognostic information regarding 1‐year mortality in symptomatic patients with severe AS undergoing TAVI. In our cohort, DC was the strongest independent predictor compared with sex, paroxysmal AF, serum Cr, and Hgb level before intervention, which were also independent predictors of death. Conventional risk predictors such as the log ES, the ES II, and LVEF were comparable in the group of survivors and nonsurvivors 1 year after TAVI. However, patients who died within 30 days after TAVI revealed a higher log ES. DC was higher in the survivor group but missed statistical significance.
Risk prediction in patients undergoing TAVI is challenging.21, 22 Previous studies showed that the conventional risk scores are inaccurate instruments to predict mid‐term mortality of aortic‐valve catheter‐based interventions.10 The log ES and ES II scores are developed and validated to predict prognosis in patients undergoing surgical aortic valve replacement and to evaluate periprocedural mortality. Notably, one study comes to the conclusion that even expert bedside evaluation seems as reliable as the risk calculation by these established scores in patients with AS.23
Hence, markers that allow for a more precise risk prediction in patients undergoing TAVI are needed. Our results demonstrate that DC is a powerful and independent predictor of 1‐year mortality in patients undergoing TAVI. Recently, we were able to demonstrate that severe autonomic failure, which is a combination of abnormal DC and abnormal heart rate turbulence, significantly predicts mortality in symptomatic patients undergoing surgical aortic valve replacement or TAVI, as well as in asymptomatic patients for whom a primarily conservative strategy was proposed.11 The results of the current study confirm the hypothesis that DC alone provides prognostic information in patients who were already intended to undergo TAVI. Due to our results, we postulate that patients at high risk identified by an impaired DC benefit from intensified peri‐interventional management including prolonged stay in an intensive care unit and heart‐rhythm monitoring. Furthermore, an impaired DC may reveal patients who do not benefit from the intervention. In these high‐risk patients, who also have a very adverse natural prognosis under conservative treatment, the risk of intervention must be carefully weighed against the outcome under conservative treatment. However, prognosis might be similar in both cases, even if previous studies revealed a mortality of 50% at 2 years and 80% at 5 years after symptom onset in patients treated conservatively.24
Mortality rate after 30 days in our study was 3.5%, which is quite low compared with other studies.25 Our results showed no significant difference of DC in survivors and nonsurvivors 30 days after TAVI. Intraprocedural adverse events leading to intrahospital death might have a major influence on short‐term outcome of our patients. These events cannot be predicted by DC. With regard to mortality after 30 days, the log ES might therefore be the better risk predictor.
Risk stratification by DC can be performed quickly and reproducibly. The calculation out of ordinary ECG recordings is done by easily manageable software. Using DC as a risk predictor is independent of the investigator, noninvasive, inexpensive, and able to preserve resources.
The pathophysiological mechanisms linking an impaired DC to mortality in patients undergoing TAVI is not known in detail. Most likely, mechanisms are similar to those that occur in heart failure, including activation of the sympathetic nervous system and the renin‐angiotensin‐aldosterone‐system.26 DC indicates the ability of the autonomic nervous system to oscillate the heart rate and to react to vagal influences.14 In patients suffering from severe AS, these mechanisms might be impaired by an overshooting sympathetic activation. DC might express the level of decompensation of neurohumoral systems and may therefore be such a strong risk predictor.
Remarkably, paroxysmal AF was shown to be a further independent risk factor for 1‐year mortality in our patients undergoing TAVI. This result is in line with previous studies of patients with AS.27 AF may be associated with a high number of comorbidities. In addition, peri‐interventional management regarding antithrombotic therapy is challenging and might trigger both bleeding or thromboembolic complications, which might lead to an increase in mortality.
4.1. Study limitations
The limitations of our study need to be mentioned. First, DC can only be assessed in patients with sufficient time in sinus rhythm to calculate the DC. Therefore, ECG recordings of patients with permanent AF had to be excluded. These patients cannot be risk‐stratified by DC, although they are known to be at high risk. Second, this study is a hypothesis‐generating study, and further prospective studies are needed to test if incorporation of DC into established risk‐assessment strategies for patients undergoing TAVI leads to a better outcome.
5. CONCLUSION
DC of heart rate is a strong and independent predictor of 1‐year‐mortality in symptomatic patients with severe AS undergoing TAVI.
Conflicts of interest
The authors declare no potential conflicts of interest.
Supporting information
Figure S1. Flowchart‐supplementary material
Duckheim M, Bensch C, Kittlitz L, et al. Deceleration capacity of heart rate predicts 1‐year mortality of patients undergoing transcatheter aortic valve implantation. DC predicts mortality of patients undergoing TAVI. Clin Cardiol. 2017;40:919–924. 10.1002/clc.22748
Funding information This study was supported by the Geschwister‐Kessel‐Stiftung, Langenenslingen, Germany
REFERENCES
- 1. Soler‐Soler J, Galve E. Worldwide perspective of valve disease. Heart. 2000;83:721–725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Iung B, Baron G, Butchart EG, et al. A prospective survey of patients with valvular heart disease in Europe: the Euro Heart Survey on Valvular Heart Disease. Eur Heart J. 2003;24:1231–1243. [DOI] [PubMed] [Google Scholar]
- 3. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation. 2002;106:3006–3008. [DOI] [PubMed] [Google Scholar]
- 4. Vahanian A, Alfieri O, Andreotti F, et al. Guidelines on the management of valvular heart disease (version 2012): the Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio‐Thoracic Surgery (EACTS). Eur J Cardiothorac Surg. 2012;42:S1–S44. [DOI] [PubMed] [Google Scholar]
- 5. Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines [published correction appears in J Am Coll Cardiol. 2014;63:2489]. J Am Coll Cardiol. 2014;63:2438–2488. [DOI] [PubMed] [Google Scholar]
- 6. Vahanian A, Alfieri O, Andreotti F, et al. Guidelines on the management of valvular heart disease (version 2012): the Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio‐Thoracic Surgery (EACTS) [article in Italian]. G Ital Cardiol (Rome). 2013;14:167–214. [DOI] [PubMed] [Google Scholar]
- 7. Roques F, Michel P, Goldstone AR, et al. The logistic EuroSCORE. Eur Heart J. 2003;24:881–882. [DOI] [PubMed] [Google Scholar]
- 8. Smith CR, Leon MB, Mack MJ, et al; PARTNER Trial Investigators . Transcatheter versus surgical aortic‐valve replacement in high‐risk patients. N Engl J Med. 2011;364:2187–2198. [DOI] [PubMed] [Google Scholar]
- 9. Vahanian A, Otto CM. Risk stratification of patients with aortic stenosis. Eur Heart J. 2010;31:416–423. [DOI] [PubMed] [Google Scholar]
- 10. D'Ascenzo F, Ballocca F, Moretti C, et al. Inaccuracy of available surgical risk scores to predict outcomes after transcatheter aortic valve replacement. J Cardiovasc Med (Hagerstown). 2013;14:894–898. [DOI] [PubMed] [Google Scholar]
- 11. Zürn CS, Rizas KD, Eick C, et al. Severe autonomic failure as a predictor of mortality in aortic valve stenosis. Int J Cardiol. 2014;176:782–787. [DOI] [PubMed] [Google Scholar]
- 12. Bauer A, Barthel P, Schneider R, et al. Improved Stratification of Autonomic Regulation for risk prediction in post‐infarction patients with preserved left ventricular function (ISAR‐Risk). Eur Heart J. 2009;30:576–583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Cygankiewicz I, Zareba W, Vazquez R, et al. Heart rate turbulence predicts all‐cause mortality and sudden death in congestive heart failure patients. Heart Rhythm. 2008;5:1095–1102. [DOI] [PubMed] [Google Scholar]
- 14. Bauer A, Kantelhardt JW, Barthel P, et al. Deceleration capacity of heart rate as a predictor of mortality after myocardial infarction: cohort study. Lancet. 2006;367:1674–1681. [DOI] [PubMed] [Google Scholar]
- 15. Eick C, Rizas KD, Meyer‐Zürn CS, et al. Autonomic nervous system activity as risk predictor in the medical emergency department: a prospective cohort study. Crit Care Med. 2015;43:1079–1086. [DOI] [PubMed] [Google Scholar]
- 16. Eick C, Rizas KD, Zürn CS, et al. Automated assessment of cardiac autonomic function by means of deceleration capacity from noisy, nonstationary ECG signals: validation study. Ann Noninvasive Electrocardiol. 2014;19:122–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lian J, Wang L, Muessig D. A simple method to detect atrial fibrillation using RR intervals. Am J Cardiol. 2011;107:1494–1497. [DOI] [PubMed] [Google Scholar]
- 18. Bauer A, Kantelhardt JW, Bunde A, et al. Phase‐rectified signal averaging detects quasi‐periodicities in non‐stationary data. Phys A Stat Mechan Appl. 2006;364:423–434. [Google Scholar]
- 19. Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41:734–745. [DOI] [PubMed] [Google Scholar]
- 20. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assn. 1958;53:457–481. [Google Scholar]
- 21. Rezq A, Basavarajaiah S, Latib A, et al. Incidence, management, and outcomes of cardiac tamponade during transcatheter aortic valve implantation: a single‐center study. JACC Cardiovasc Interv. 2012;5:1264–1272. [DOI] [PubMed] [Google Scholar]
- 22. Urena M, Webb JG, Tamburino C, et al. Permanent pacemaker implantation after transcatheter aortic valve implantation: impact on late clinical outcomes and left ventricular function. Circulation. 2014;129:1233–1243. [DOI] [PubMed] [Google Scholar]
- 23. Laurent M, Fournet M, Feit B, et al. Simple bedside clinical evaluation versus established scores in the estimation of operative risk in valve replacement for severe aortic stenosis. Arch Cardiovasc Dis. 2013;106:651–660. [DOI] [PubMed] [Google Scholar]
- 24. Otto CM. Timing of aortic valve surgery. Heart. 2000;84:211–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Hamid T, Choudhury TR, Anderson SG, et al. Does the CHA2DS2‐Vasc score predict procedural and short‐term outcomes in patients undergoing transcatheter aortic valve implantation? Open Heart. 2015;2:e000170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Coats AJ. The importance and complexity of neurohumeral over‐activity in chronic heart failure. Int J Cardiol. 2000;73:13–14. [DOI] [PubMed] [Google Scholar]
- 27. Sannino A, Gargiulo G, Schiattarella GG, et al. A meta‐analysis of the impact of pre‐existing and new‐onset atrial fibrillation on clinical outcomes in patients undergoing transcatheter aortic valve implantation. EuroIntervention. 2016;12:e1047–e1056. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Flowchart‐supplementary material