Little is known about the usefulness of evaluating cardiac neurohormones in patients undergoing transcatheter aortic valve implantation (TAVI). The objectives of this study were to evaluate the baseline values and serial changes of N-terminal prohormone B-type natriuretic peptide (NT-proBNP) after TAVI, its related factors, and prognostic value. A total of 333 consecutive patients were included, and baseline, procedural, and follow-up (median 20 months, interquartile range 9 to 36) data were prospectively collected. Systematic NT-proBNP measurements were performed at baseline, hospital discharge, 1, 6, and 12 months, and yearly thereafter. Baseline NT-proBNP values were elevated in 86% of the patients (median 1,692 pg/ml); lower left ventricular ejection fraction and stroke volume index, greater left ventricular mass, and renal dysfunction were associated with greater baseline values (p <0.01 for all). Higher NT-proBNP levels were independently associated with increased long-term overall and cardiovascular mortalities (p <0.001 for both), with a baseline cut-off level of ∼2,000 pg/ml best predicting worse outcomes (p <0.001). At 6- to 12-month follow-up, NT-proBNP levels had decreased (p <0.001) by 23% and remained stable up to 4-year follow-up. In 39% of the patients, however, there was a lack of NT-proBNP improvement, mainly related to preprocedural chronic atrial fibrillation, lower mean transaortic gradient, and moderate-to-severe mitral regurgitation (p <0.01 for all). In conclusion, most patients undergoing TAVI presented high NT-proBNP levels, and a lack of improvement was observed in >1/3 of the patients after TAVI. Also, higher NT-proBNP levels predicted greater overall and cardiac mortalities at a median follow-up of 2 years. These findings support the implementation of NT-proBNP measurements for the clinical decision-making process and follow-up of patients undergoing TAVI.
Natriuretic peptides (NPs) are elevated in a number of cardiovascular diseases such as cardiac hypertrophy, acute coronary syndromes, and heart failure. In the context of aortic stenosis (AS), both B-type natriuretic peptide (BNP) and its prohormone—N-terminal prohormone B-type natriuretic peptide (NT-proBNP)—levels are also elevated, and the degree of their increase has been correlated with the severity of AS, symptom status, and clinical outcomes after surgical aortic valve replacement (SAVR). In the context of transcatheter aortic valve implantation (TAVI), some studies have suggested an association between NP levels and early and 1-year outcomes after TAVI. However, most of these studies included a limited number of patients, relatively short follow-up periods (≤1 year), and very few data on cardiovascular outcomes (i.e., cardiac death, heart failure). More importantly, although a significant decrease in NP levels has been shown after TAVI, the degree of these changes and the factors associated with the lack of cardiac neurohormonal improvement have not yet been evaluated. The aim of this study was therefore to evaluate the serial changes, related factors, and the prognostic significance of NT-proBNP on the long-term follow-up of a large cohort of patients undergoing TAVI.
Methods
A total of 333 consecutive patients with symptomatic AS considered as not suitable or at very high risk for SAVR underwent a TAVI procedure and were included in the study. Details about the TAVI procedure have been provided elsewhere. All baseline and procedural characteristics were prospectively collected on preset data collection forms. Baseline co-morbidities were defined according to the Society of Thoracic Surgeons criteria, and periprocedural events were defined according to the Valve Academic Research Consortium-2 criteria. Coronary artery disease was defined as the presence of coronary lesion with a diameter stenosis of ≥50% in vessels ≥2.0 mm or previous coronary revascularization. The procedures were performed under a compassionate Clinical Program approved by the Health Canada, and all patients provided signed informed consent for the procedures.
Clinical follow-up was carried out at 30 days, 6 and 12 months, and yearly thereafter. The median follow-up was 20 months (interquartile range 9 to 36), and no patient was lost to follow-up. All clinical events during the follow-up period were defined according to the Valve Academic Research Consortium-2 criteria. All patients underwent a Doppler echocardiographic examination at baseline before the intervention and at hospital discharge. The New York Heart Association (NYHA) class was evaluated before the procedure and at each time point during the follow-up period. Plasma NT-proBNP levels were measured within 48 hours before the TAVI procedure and thereafter daily during hospitalization, at 30 days, 6 months, 1 year, and yearly thereafter. The blood samples were collected in EDTA-containing tubes and were immediately centrifugated. The NT-proBNP level was measured using a chemoluminescent immunoassay kit (Elecsys proBNP II; Roche, Minneapolis, Minnesota; a normal value of <450 pg/ml for the general population). The increase in NT-proBNP levels was also evaluated using the suggested cut-off levels for elderly (normal values of <1,800 and <900 pg/ml for ages >75 years and from 50 to 75 years, respectively) and renal dysfunction (<1,200 pg/ml). At 6- to 12-month follow-up, the patients were considered as nonresponders (failure to improve their NT-proBNP levels) if their NT-proBNP value was equal or greater compared with the baseline value.
The following measurements were obtained in all patients: aortic annulus diameter, left ventricular ejection fraction (LVEF) calculated with the Simpson method, mean transvalvular gradient calculated with the Bernoulli formula, and the valve effective orifice area (EOA) measured by the continuity equation. The EOA was indexed (EOAi) for body surface area, and the occurrence of prosthesis-patient mismatch was defined as severe if the EOAi was <0.65 cm 2 /m 2 . In patients with a body mass index of ≥30 kg/m 2 , the prosthesis-patient mismatch was classified as severe if the EOAi was <0.60 cm 2 /m 2 . The presence and degree of aortic regurgitation was recorded in all patients. The degree of aortic regurgitation was classified as follows: trivial, mild, moderate, and severe. The left ventricular mass (LVM) was calculated as follows: LVM (g) = 1.04 ([LV end-diastolic diameter + ventricular septal thickness + posterior wall thickness] 3 − [LV end-diastolic diameter] 3 ) − 13.6. Stroke volume was obtained by multiplying the left ventricular outflow tract area by the velocity-time integral measured by pulsed wave Doppler in the left ventricular outflow tract. Both the LVM and stroke volume indexes were calculated in relation to the body surface area.
Continuous variables are expressed as mean (SD) or median (25% to 75% interquartile range) according to the variable distribution. Group comparisons were analyzed using the Student t or Wilcoxon rank-sum test for continuous variables and chi-square or Fisher’s exact test for categorical variables. A repeated measures model with interaction was used to analyze the changes of NT-proBNP over time. The model including log-transformed NT-proBNP satisfied the assumption of normality of residuals. Posterior comparisons were performed using the Tukey test. The predictors of increased NT-proBNP values at baseline were determined using a linear regression analysis, and the lack of NT-proBNP improvement at 1-year follow-up was determined using a logistic regression analysis. A univariate logistic regression analysis was used to determine the predictors of 30-day mortality. Univariate and multivariate Cox proportional hazards models were used to determine the predictors of cumulative late mortality, cardiac mortality, and the composite of cardiac mortality and rehospitalization due to heart failure. All the variables exhibiting a p value of <0.05 at the univariate analysis were included in the multivariate model. A receiver operating characteristic (ROC) curve analysis was used to determine the best baseline NT-proBNP cut-off levels predicting increased overall late mortality, cardiac mortality, and the composite of cardiac mortality and rehospitalizations due to heart failure at follow-up. Survival curves were constructed using Kaplan-Meier estimates, and the log-rank test was used for comparison between groups. The results were considered significant with p values of <0.05. All analyses were conducted using the statistical package SAS, version 9.3 (SAS Institute Inc., Cary, North Carolina).
Results
The main baseline and procedural characteristics of the study population are listed in Table 1 . Baseline NT-proBNP levels were elevated in 86% of the patients using the standard criteria (450 pg/ml), with a median baseline NT-proBNP value of 1,692 pg/ml (interquartile range 667 to 3,910). Also, 75.8% and 53.2% of the patients had increased NT-proBNP levels according to the criteria of Kim et al for renal failure and age, respectively. The distribution of the study population according to the baseline NT-proBNP levels is shown in Figure 1 . The factors associated with greater baseline NT-proBNP levels are listed in Table 2 . In the multivariate analysis, the variables associated with greater NT-proBNP levels were renal dysfunction (R = 0.097, p <0.001), lower LVEF (R 2 = 0.127, p <0.001), lower stroke volume index (R 2 = 0.015, p = 0.027), and greater LVM index (R 2 = 0.033, p <0.001). A total of 116 patients (34.8%) had died after a median follow-up of 20 months (interquartile range 9 to 36), 61 of them (18.3%) from cardiac causes. The variables associated with a greater risk of cumulative late mortality are listed in Table 3 . The baseline NT-proBNP level independently predicted an increased late mortality (hazard ratio 1.03 for each increase of 1,000 pg/ml, 95% confidence interval [CI] 1.01 to 1.08), and it was the only independent predictor of both cardiac mortality (hazard ratio 1.04 for each increase of 1,000 pg/ml, 95% CI 1.01 to 1.08) and the combined end point of cardiac mortality and rehospitalization due to heart failure after TAVI (hazard ratio 1.03 for each increase of 1,000 pg/ml, 95% CI 1.01 to 1.09). A baseline NT-proBNP cut-off value of 1,900 pg/ml best identified the patients at greater risk for late cumulative mortality (area under the ROC curve of 0.65 [95% CI: 0.59 to 0.71], sensitivity 60.9%, specificity 59.7%, p <0.001). A baseline NT-proBNP cut-off value of 2,200 pg/ml best identified the patients at greater risk for cardiac death or cardiac death/rehospitalization due to heart failure (area under the ROC curve of 0.64 [95% CI: 0.58 to 0.71], sensitivity 58.3%, specificity 64.6%, p <0.001). The Kaplan-Meier survival curves according to the baseline NT-proBNP values (< or ≥2,000 pg/ml) are shown in Figure 2 .
| Variable | All Patients (n = 333) |
|---|---|
| Clinical variable | |
| Age (yrs) | 79.6 ± 7.8 |
| Men | 177 (53.2) |
| NYHA functional class | |
| I–II | 71 (21.3) |
| III–IV | 262 (78.7) |
| Diabetes mellitus | 113 (33.9) |
| Hypertension | 293 (88.0) |
| Coronary artery disease | 210 (63.1) |
| Previous coronary artery bypass graft | 126 (37.8) |
| Atrial fibrillation | 101 (30.3) |
| Cerebrovascular disease | 65 (19.5) |
| Peripheral vascular disease | 116 (34.8) |
| Chronic obstructive pulmonary disease | 99 (29.7) |
| Estimated glomerular filtration (ml/min) | 57.4 ± 23.0 |
| Society of Thoracic Surgeons–predicted risk of mortality (%) | 7.3 ± 4.9 |
| NT-proBNP (pg/ml) | 1,692 (667–3,910) |
| Echocardiographic variable before procedure | |
| Left ventricular ejection fraction (%) | 53.8 ± 13.8 |
| Mean aortic gradient (mm Hg) | 41.0 ± 16.1 |
| Aortic valve area (cm 2 ) | 0.65 ± 0.21 |
| Pulmonary systolic arterial pressure (mm Hg) | 43.2 ± 14.0 |
| Moderate-to-severe mitral regurgitation | 86 (25.8) |
| Stroke volume index (ml/m 2 ) | 35.1 ± 10.2 |
| LVM index (g/m 2 ) | 117.8 ± 36.0 |
| Procedural variable | |
| Procedural success ∗ | 285 (85.6) |
| Approach | |
| Transapical | 177 (53.2) |
| Transfemoral | 131 (39.3) |
| Transaortic | 25 (7.5) |
| Prosthesis type | |
| Sapien | 188 (56.5) |
| Sapien XT | 131 (39.3) |
| Sapien 3 | 7 (2.1) |
| Portico | 7 (2.1) |
| Prosthesis size (mm) | |
| 20 | 2 (0.6) |
| 23 | 174 (52.7) |
| 26 | 128 (38.8) |
| 29 | 26 (7.9) |
| Echocardiographic variable after procedure | |
| Left ventricular ejection fraction (%) | 53.4 ± 13.0 |
| Mean aortic gradient (mm Hg) | 12.0 ± 6.7 |
| Aortic valve area (cm 2 ) | 1.46 ± 0.35 |
| Moderate-to-severe aortic regurgitation | 40 (13.7) |
| Severe prosthesis-patient mismatch | 37 (14.4) |
| Variable | Univariate | Multivariate Model | ||||
|---|---|---|---|---|---|---|
| Standardized Coefficient (β) | R 2 | p Value | Standardized Coefficient (β) | R 2 | p Value | |
| Clinical variable | ||||||
| Age (yrs) | −0.019 | 0.0004 | 0.719 | |||
| Men | 0.107 | 0.0114 | 0.058 | |||
| NYHA functional class III–IV | 0.089 | 0.0081 | 0.103 | |||
| Diabetes mellitus | 0.074 | 0.0055 | 0.177 | |||
| Hypertension | 0.002 | 0.00001 | 0.971 | |||
| Coronary artery disease | 0.104 | 0.0109 | 0.058 | |||
| Previous coronary artery bypass graft | −0.007 | 0.00001 | 0.901 | |||
| Chronic atrial fibrillation | 0.043 | 0.0019 | 0.434 | |||
| Cerebrovascular disease | 0.094 | 0.0089 | 0.086 | |||
| Peripheral vascular disease | 0.048 | 0.0023 | 0.381 | |||
| Chronic obstructive pulmonary disease | −0.059 | 0.0035 | 0.282 | |||
| Estimated glomerular filtration (ml/min) | −0.333 | 0.111 | <0.001 | −0.282 | 0.097 | <0.001 |
| Society of Thoracic Surgeons–predicted risk of mortality (%) | 0.242 | 0.058 | <0.001 | |||
| Echocardiographic variable | ||||||
| Left ventricular ejection fraction (%) | −0.328 | 0.108 | <0.001 | −0.213 | 0.127 | <0.001 |
| Mean aortic gradient (mm Hg) | −0.036 | 0.001 | 0.514 | |||
| Aortic valve area (cm 2 ) | −0.093 | 0.009 | 0.096 | |||
| Pulmonary systolic arterial pressure (mm Hg) | 0.140 | 0.019 | 0.015 | 0.089 | 0.008 | 0.108 |
| Moderate-to-severe mitral regurgitation | 0.069 | 0.005 | 0.211 | |||
| Stroke volume index (ml/m 2 ) | −0.166 | 0.027 | 0.005 | −0.136 | 0.015 | 0.027 |
| Left ventricular mass index (g/m 2 ) | 0.316 | 0.100 | <0.001 | 0.226 | 0.033 | 0.001 |
| Variable | Univariate | p Value | Multivariate Model | p Value |
|---|---|---|---|---|
| OR/HR (95% CI) | HR (95% CI) | |||
| 30-Day mortality (n = 29) | ||||
| Men | 2.24 (1.01–5.04) | 0.049 | — | — |
| Baseline NT-proBNP ∗ | 1.04 (1.01–1.08) | 0.043 | — | — |
| Cumulative mortality (n = 116) | ||||
| Men | 1.64 (1.14–2.38) | 0.008 | — | — |
| Chronic atrial fibrillation | 1.57 (1.03–2.38) | 0.034 | — | — |
| Chronic obstructive pulmonary disease | 1.65 (1.14–2.39) | 0.009 | 1.72 (1.11–2.66) | 0.015 |
| Estimated glomerular filtration † | 1.20 (1.08–1.33) | 0.001 | 1.18 (1.03–1.36) | 0.026 |
| Baseline NT-proBNP ∗ | 1.07 (1.02–1.12) | <0.001 | 1.03 (1.01–1.08) | 0.045 |
| Mean aortic gradient ‡ | 1.15 (1.01–1.33) | 0.033 | — | — |
| Stroke volume index § | 1.35 (1.04–1.75) | 0.003 | 1.27 (1.01–1.68) | 0.034 |
| Cumulative cardiac mortality (n = 61) | ||||
| Baseline NT-proBNP ∗ | 1.06 (1.02–1.10) | <0.001 | 1.04 (1.01–1.08) | 0.035 |
| Left ventricular ejection fraction ‖ | 1.12 (1.02–1.23) | 0.019 | — | — |
| Moderate-to-severe mitral regurgitation | 1.72 (1.01–2.93) | 0.046 | — | — |
| Stroke volume index § | 1.39 (1.01–1.92) | 0.013 | — | — |
| Cumulative noncardiac mortality (n = 55) | ||||
| Men | 1.77 (1.03–3.02) | 0.038 | — | — |
| Chronic atrial fibrillation | 1.86 (1.04–3.33) | 0.037 | — | — |
| Chronic obstructive pulmonary disease | 1.82 (1.07–3.12) | 0.029 | 2.11 (1.20–3.73) | 0.009 |
| Estimated glomerular filtration † | 1.18 (1.03–1.35) | 0.047 | 1.20 (1.04–1.39) | 0.008 |
| Cumulative cardiac mortality and/or rehospitalization for heart failure (n = 96) | ||||
| Men | 1.52 (1.01–2.27) | 0.044 | — | — |
| Chronic atrial fibrillation | 1.91 (1.24–2.95) | 0.004 | — | — |
| Estimated glomerular filtration † | 1.16 (1.04–1.30) | 0.020 | — | — |
| Baseline NT-proBNP ∗ | 1.05 (1.01–1.09) | 0.002 | 1.03 (1.01–1.09) | 0.026 |
| Left ventricular ejection fraction ‖ | 1.15 (1.06–1.25) | 0.002 | — | — |
| Mean aortic gradient ‡ | 1.27 (1.08–1.50) | 0.002 | 0.98 (0.97–0.99) | 0.340 |
| Moderate-to-severe mitral regurgitation | 1.59 (1.04–2.45) | 0.034 | — | — |
| Stroke volume index § | 1.35 (1.03–1.77) | 0.007 | — | — |
| Left ventricular mass index ¶ | 1.01 (1.01–1.01) | 0.037 | — | — |
∗ Per 1,000 pg/ml increase in N-terminal B-type natriuretic peptide.
† Per 10 ml/min decrease in estimated glomeruar filtration rate.
‡ Per 10-mm Hg decrease in mean aortic gradient.
§ Per 10-ml decrease in stroke volume index.
‖ Per 5% decrease in left ventricular ejection fraction.
The early and late changes in NT-proBNP levels after TAVI are shown in Figure 3 . The NT-proBNP levels decreased by −23% (−62; +31) at 1-year follow-up and remained stable up to 4-year follow-up (p <0.001). The early changes in NT-proBNP levels differed according to the approach used during the TAVI procedure. In patients treated by transfemoral approach, the NT-proBNP levels did not change immediately after the procedure (hospital discharge, p = 0.799) and decreased up to 6- to 12-month follow-up (−25% [−69; +37], p <0.001), whereas in patients treated by transapical or transaortic approach, the NT-proBNP levels increased at hospital discharge (+23% [−20; +127] in the transapical group, p <0.001; +32% [+23; +146] in the transaortic group, p = 0.007), decreased afterward until 6- to 12-month follow-up, and then remained stable up to 4 years (p <0.001 and p = 0.003, for the transapical and transaortic groups, respectively).
At 6- to 12-month follow-up, a total of 69 patients (of 179 patients at risk, 39%) failed to improve their NT-proBNP levels compared with baseline. The changes in NT-proBNP levels between baseline and 1-year follow-up in these patients (nonresponders) compared with those who had improved their NT-proBNP levels (responders) are shown in Figure 4 , and the main baseline and echocardiographic characteristics of these 2 groups are compared in Table 4 . Patients in the nonresponder group increased the NT-proBNP values by +71% (+22; +164) at 1-year follow-up (p <0.001) compared with a decrease of −51% (−75; −31) in the responder group (p <0.001), p <0.001 for comparison between groups. The factors associated with the lack of NT-proBNP improvement after TAVI are listed in Table 5 . In the multivariate analysis, the predictors of the lack of NT-proBNP improvement were chronic atrial fibrillation (AF; odds ratio 2.40, 95% CI 1.06 to 5.44, p = 0.036), a lower mean gradient (odds ratio 0.98, 95% CI 0.95 to 0.99 per 10 mm Hg, p = 0.025), and moderate-to-severe mitral regurgitation (MR; odds ratio 2.11, 95% CI 1.03 to 4.34, p = 0.042).
| Variable | Nonresponder (n = 69) | Responder (n = 109) | p Value |
|---|---|---|---|
| Clinical variable | |||
| Age (yrs) | 78.8 ± 7.3 | 80.5 ± 7.3 | 0.123 |
| Men | 40 (58.0) | 68 (62.4) | 0.637 |
| NYHA class | |||
| I–II | 17 (24.6) | 17 (15.6) | 0.171 |
| III–IV | 52 (75.4) | 92 (84.4) | |
| Diabetes mellitus | 24 (34.8) | 37 (33.9) | 1.00 |
| Hypertension | 60 (87.0) | 98 (89.9) | 0.628 |
| Coronary artery disease | 38 (55.1) | 69 (63.3) | 0.346 |
| Previous coronary artery bypass graft | 28 (40.6) | 36 (33.0) | 0.338 |
| Atrial fibrillation | 20 (29.0) | 15 (13.8) | 0.019 |
| Cerebrovascular disease | 12 (17.4) | 23 (21.1) | 0.568 |
| Peripheral vascular disease | 29 (42.0) | 34 (31.2) | 0.151 |
| Chronic obstructive pulmonary disease | 19 (27.5) | 25 (22.9) | 0.593 |
| Estimated glomerular filtration (ml/min) | 56.5 ± 22.4 | 56.8 ± 22.9 | 0.932 |
| Society of Thoracic Surgeons–predicted risk of mortality (%) | 6.8 ± 4.1 | 7.2 ± 4.1 | 0.522 |
| Procedural success ∗ | 57 (82.6) | 98 (89.9) | 0.174 |
| Approach | |||
| Transfemoral | 22 (31.9) | 45 (41.3) | 0.472 |
| Transapical | 44 (63.8) | 59 (54.1) | |
| Transaortic | 3 (4.3) | 5 (4.6) | |
| Echocardiographic variable before procedure | |||
| Left ventricular ejection fraction (%) | 55.1 ± 12.6 | 53.3 ± 14.6 | 0.417 |
| Mean gradient (mm Hg) | 36.6 ± 11.7 | 43.3 ± 18.7 | 0.009 |
| Aortic valve area (cm 2 ) | 0.68 ± 0.16 | 0.63 ± 0.24 | 0.185 |
| Pulmonary systolic arterial pressure (mm Hg) | 43.6 ± 14.6 | 43.0 ± 13.2 | 0.799 |
| Moderate-to-severe mitral regurgitation | 23 (33.3) | 22 (20.2) | 0.049 |
| Stroke volume index (ml/m 2 ) | 37.0 ± 7.8 | 35.9 ± 11.0 | 0.481 |
| Left ventricular mass index (g/m 2 ) | 118.3 ± 36.0 | 119.4 ± 32.7 | 0.836 |
| Echocardiographic variable after procedure | |||
| Left ventricular ejection fraction (%) | 55.3 ± 12.5 | 54.1 ± 12.8 | 0.559 |
| Mean gradient (mm Hg) | 11.4 ± 6.6 | 12.5 ± 6.13 | 0.262 |
| Aortic valve area (cm 2 ) | 1.48 ± 0.38 | 1.42 ± 0.32 | 0.303 |
| Aortic regurgitation ≥2 | 16 (23.2) | 20 (18.3) | 0.449 |
| Severe prosthesis mismatch | 8 (11.6) | 12 (11.0) | 1.00 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
