In patients with aortic stenosis (AS), the relation of cardiac energetic pathways with cardiac structure and function, their changes, and their prognostic significance are not well understood. We aimed to characterize metabolic profiles in patients with severe AS before and after aortic valve replacement (AVR) and their association with functional status, structural remodeling and mortality. Patients with symptomatic, severe AS before ( n = 143) and 1-year after ( n = 113) AVR underwent cardiac magnetic resonance (CMR), serum cardiac biomarkers, and 6-minute walk test. Resting nonfasting plasma samples underwent targeted nuclear magnetic resonance (NMR) for fatty acids (FA), branched chain amino acids (BCAAs), glycolysis-related metabolites, and ketones. Lower FA and BCAA concentrations, but not glycolysis metabolites or ketones, correlated with greater myocardial mass and focal fibrosis, NT-proBNP, TnT and 6-minute walk distance. After 10.5 years of follow-up (66/143 deaths), lower FAs and BCAAs, but not ketones were independently associated with higher mortality risk (p <0.05). At 1-year after AVR, FAs had decreased compared to baseline. In conclusion, reduced serum FA and BCAA concentrations are cardiac, maladaptive, prognostic metabolic changes to AS, which are not reversible after AVR. Whether these markers may be used to guide the timing of AVR or provide metabolic risk stratification remains to be evaluated by future research. In patients with AS systemic metabolomics and their association with myocardial remodeling and outcome after AVR are largely unknown. We show that in severe AS low levels of unsaturated FA and BCAAs correlate with higher mortality risk, and biomarkers measured by CMR, serum, and functional incapacity, and do not increase after AVR. This may provide an alternate approach to risk stratification using blood biomarkers or guide targeted therapies to myocardial energetics before or after AVR.
Graphical abstract
Metabolites in severe aortic stenosis and their associations with cardiac remodeling/damage and mortality after aortic valve replacement (AVR). FFAs indicates free fatty acids; BCAA, branched chained amino acids.
In aortic stenosis (AS) progressive narrowing of the aortic valve causes pressure overload of the left ventricle, myocardial remodeling and heart failure. Structural alterations of the myocardium are accompanied by changes in myocardial energy metabolism. Whereas the normal heart predominantly uses fatty acids (FA) as energy source (60-90%), , substrate utilization is altered in AS with less FA oxidation and increased glucose use. , It is unclear whether changes in myocardial metabolism are adaptive or maladaptive, and whether they are reversible. High throughput metabolomic profiling permits systematic assessment of metabolomic pathways related to the heart. , Serum changes may reflect systemic adaptation to the hemodynamic consequences of AS, or intrinsic myocardial metabolic remodeling. Linkage of metabolic alterations to functional, structural, and biochemical markers of myocardial injury and their trajectories after intervention may help to untangle this relationship and provide novel risk measures that are easily sampled by blood tests. A more nuanced understanding of metabolic derangement in AS may be of substantial clinical interest as therapeutic strategies aimed at manipulation of cardiac substrate utilization have not proven to be of substantial benefit, and untangling the interplay between cardiac structural changes and metabolism may yield pathophysiological insights. We hypothesized that plasma metabolic markers in AS would: (1) associate with imaging, biochemical and functional cardiovascular disease markers, (2) normalize after valve replacement and (3) portend prognostic value.
Methods
Study population
In a prospective observational cohort study, patients with severe symptomatic AS who underwent AVR between January 2012 and January 2015 were included at a single tertiary referral cardiac center, University College London Hospital NHS Trust, London, United Kingdom. The study was approved by the ethical committee of the U.K. National Research.
Ethics Service (07/H0715/101) and this work is a substudy of the previously published RELIEF-AS study (Regression of Myocardial Fibrosis After Aortic Valve Replacement; NCT02174471). The study conformed to the principles of the Helsinki Declaration, and all subjects gave written informed consent. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research. Patients were recruited before preoperative evaluation. Pre-AVR and post-AVR, a comprehensive assessment included clinical history, blood pressure, 6-min walk test (6MWT), blood sampling (for high-sensitivity troponin T [hsTnT] and N-terminal pro–B-type natriuretic peptide [NT-proBNP]), electrocardiography, transthoracic echocardiography, and cardiac magnetic resonance (CMR) using the same equipment. Non fasting blood was collected, and plasma was separated by centrifugation and stored at −80 degrees Celsius. Inclusion and exclusion criteria for the study have previously been described. In short, only patients with severe AS and clear indication for AVR, no greater than moderate valve disease other than AS, and ability to undergo CMR were included.
Study procedures
Metabolic marker profiling
A targeted nuclear magnetic resonance (NMR) metabolomics platform was used for the quantification of metabolites in serum samples. ,,, Metabolite assessment was performed at baseline (pre-AVR) and follow-up (at 1-year after AVR). Overall, out of >150 metabolites measured, 21 metabolites were selected as reflecting the main energetic pathways related to the heart: FAs, amino acids, glucose, lactate, and ketone bodies. FA composition included saturated, monounsaturated (MUFA) and polyunsaturated (PUFA) FAs. Glycolytic metabolites (glucose, citrate) were measured in relation to glucose metabolisms. Amino acids (AAs) were categorized into branched-chained AAs, levels of which have been shown to undergo changes in pressureoverloaded mouse models, and nonbranched-chained AAs. Acetate and 3-hydroxybutyrate were measured as ketone bodies. The reliability of assays was assessed in a sample of 12 patients who underwent repeat analysis on a second serum sample. This showed excellent concordance (intra class correlation coefficient 0.92-0.99 for all metabolites analyzed).
Multimodality cardiac imaging
Echocardiography was primarily used to assess ventricular function, diastolic parameters and valve area or velocities. Contrast-enhanced CMR was performed for deep structural and functional phenotyping as previously reported, and included cine imaging, quantitative late gadolinium enhancement (LGE), T1 mapping and extracellular volume fraction. LGE was quantified with the 3-standard deviation above normal myocardium method and calculated as percentage of the LV. All analysis were performed by operators blinded to clinical parameters. CMR image analysis was done using CVI42 software (version 5.1.2[303], Circle Cardiovascular Imaging, Calgary, Alberta, Canada).
Outcome analysis
All-cause mortality was captured from the National Health Service Spine database and was 100% complete.
Statistical analysis
All continuous variables are expressed as mean ± standard deviation (SD) or median (interquartile range [IQR]) for skewed data. Normality was checked using the Shapiro-Wilk test. Categorical variables are expressed as percentages. Groups were compared using the Mann-Whitney U test or test chi-square test, as appropriate. Changes between pre-AVR and post-AVR visits were compared using the paired t-test and McNemar’s test, as appropriate.
To identify associations with quantitative markers of cardiac structure (LVMi), cardiac damage (TnT, NT-proBNP, extent of focal scar on CMR), and functional capacity (6-MWT) and their longitudinal changes, multiple linear regression analysis was performed for all metabolomic markers. Adjustment of baseline models was performed for potential confounders (age, sex, BMI, arterial hypertension, diabetes, eGFR), which were chosen based on prior knowledge. A false discovery rate of 0.1 was used to determine significant associations. To test whether baseline serum metabolites explain cardiac reverse remodeling and functional recovery (i.e., the change in TnT, NT-proBNP, 6-MWT), regression models were adjusted for age, sex, BMI, arterial hypertension, diabetes, eGFR, delta mean transvalvular gradient, and the baseline value of the respective marker. The association of metabolite components with all-cause mortality was assessed using multivariate Cox proportional hazard regression models with adjustment performed for potential confounders (age, sex, hypertension, diabetes). Univariate models are also presented. A 2-sided p-value of <0.05 was considered significant. Statistical analyses were carried out using SPSS software version 28 (IBM, Armonk, New York).
Results
Patient characteristics
In total, 181 people with severe AS were screened and 38 were excluded due to various reasons ( Figure 1 ). The remaining 143 patients with both metabolomic and multimodality imaging data formed the baseline AS cohort. Of these 143 AS patients, 113 returned for 1-year follow-up with repeat metabolomic and imaging ( n = 106, post-AVR pacemaker in 7 patients with only metabolomic evaluation at follow-up) assessment. Detailed patient characteristics are displayed in Tables 1 and 2 .
Patient flow chart. AS = aortic stenosis; AVR = aortic valve replacement; CMR = cardiac magnetic resonance imaging.
Table 1
Patient baseline characteristics
| Severe aortic stenosis n = 143 | |
|---|---|
| Demographics | |
| Age, years | 72 (65-77) |
| Male sex, % | 55.2 |
| BMI, kg/m 2 | 28.0 (24.9-31.5) |
| Ethnicity, % | |
| European | 93.8 |
| South Asian | 4.7 |
| African Caribbean | 1.5 |
| EuroSCORE-II, % | 1.5 (1.0-2.5) |
| Clinical parameters | |
| Angina, % | 36.6 |
| Syncope, % | 9.2 |
| NYHA functional class, % | |
| I | 13.2 |
| II | 49.3 |
| III | 34.6 |
| IV | 2.9 |
| 6-minute walk distance, m | 480 (333-584) |
| Diabetes, % | 19.7 |
| Hypertension, % | 80.3 |
| Atrial fibrillation, % | 16.1 |
| Coronary artery disease, % | 31.0 |
| Laboratory results | |
| NT-proBNP, pg/mL | 76 (30-251) |
| Troponin, pg/mL | 14 (10-20) |
| eGFR, mL/min | 73 (60-92) |
| Imaging parameters | |
| Echocardiography | |
| Mean gradient, mmHg | 46 (38-55) |
| CMR | |
| Vmax, m/s | 3.8 (3.3-4.2) |
| LVEF, % | 73 (64-80) |
| LVEDV, mL | 118 (94-147) |
| LVESV, mL | 29 (19-49) |
| LV stroke volume, mL | 82 (69-95) |
| LV mass index, g/m2 | 83 (70-106) |
| LGE mass*, % | 9.2 (5.0-17.5) |
| T1 relaxation times precontrast, ms | 1048 (1019-1073) |
| ECV, % | 28 (27-30) |
| Procedural characteristics | |
| SAVR, % | 97.2 |
| Tissue prosthesis | 68.5 |
| Mechanical prosthesis | 25.9 |
| Sutureless prosthesis | 2.8 |
| TAVI, % | 2.8 |
Continuous parameters are shown as median and interquartile ranges and categorical as percentages.
*LGE mass (percent) as the median of all patients, including those without LGE.
BMI = body mass index; ECV = extracellular volume fraction; eGFR = estimated glomerular filtration rate; LGE = late gadolinium enhancement; LVEDV = left ventricular end-diastolic volume; LVEF = left ventricular ejection fraction; LVESV = left ventricular end-systolic volume; NT-proBNP = N-terminal pro-brain natriuretic peptide; NYHA = New York Heart Association functional class; Vmax = maximum transaortic velocity.
Table 2
Baseline metabolomics
| Severe aortic stenosis n = 143 | |
|---|---|
| Glucose metabolism | |
| Glucose, mmol/L | 4.9 (4.1-6.1) |
| Lactate, mmol/L | 1.2 (1.0-1.6) |
| Citrate, mmol/L | 0.19 (0.17-0.21) |
| Fatty acids | |
| Total fatty acids, mmol/L | 10.8 (8.8-12.6) |
| Estimated degree of unsaturation | 1.24 (1.19-1.28) |
| Omega-3 fatty acids, mmol/L | 4.1 (3.1-5.1) |
| 22:6, docosahexaenoic acid, mmol/L | 0.13 (0.10-0.17) |
| Omega-6 fatty acids, mmol/L | 3.4 (2.8-4.0) |
| 18:2, linoleic acid, mmol/L | 2.8 (2.2-3.3) |
| Polyunsaturated fatty acids, mmol/L | 3.9 (3.1-4.5) |
| Monounsaturated fatty acids; 16:1, 18:1, mmol/L | 2.9 (2.2-3.4) |
| Saturated fatty acids, mmol/L | 4.2 (3.5-4.7) |
| Amino acids (AA) | |
| Nonbranched chained AA | |
| Alanine, mmol/L | 0.40 (0.36-0.45) |
| Glutamine, mmol/L | 0.46 (0.40-0.53) |
| Histidine, mmol/L | 0.05 (0.05-0.06) |
| Branched chained AA | |
| Isoleucine, mmol/L | 0.06 (0.05-0.08) |
| Leucine, mmol/L | 0.08 (0.07-0.10) |
| Valine, mmol/L | 0.16 (0.13-0.18) |
| Aromatic AA | |
| Phenylalanine, mmol/L | 0.06 (0.06-0.07) |
| Tyrosine, mmol/L | 0.06 (0.05-0.07) |
| Ketone bodies | |
| Acetoacetate, mmol/L | 0.05 (0.05-0.06) |
| 3-hydroxybutyrate, mmol/L | 0.11 (0.10-0.14) |
Association of metabolic markers and cardiac structure, functional capacity, troponin-T and NT-proBNP
Full univariate and multivariate models are presented in the supplement (Supplementary Table S1). Lower FA levels were independently associated with more LV remodeling (higher LVMi), scar (higher amount of LGE), and damage (higher NT-proBNP and hs-TnT), and worse functional capacity (lower 6-minute walk distance). Lower levels of branched-chained AAs (BCAA) were linked to increased hypertrophic remodeling and cardiac damage, as indicated by significant associations of Isoleucin and Leucin with LVMi (p = 0.001 and 0.002), NT-proBNP (both p <0.001) and hs-TnT (p = 0.025 and 0.013). Also, valine was significantly associated with NT-proBNP (p = 0.005) and hs-TnT (p = 0.016).
Changes in metabolic markers after AVR
At 1-year after AVR, symptoms, valve gradients, structural LV remodeling, and markers of cardiac injury and decompensation improved ( Table 3 ). Levels of polyunsaturated FA in AS had decreased at 1-year post-AVR (e.g., Omega-6 FA: 3.5 to 3.2 mmol/l; Omega-3 FA: 4.1 to 3.9 mmol/l, both p <0.05). AA and ketone body concentrations had not changed at follow-up (all p >0.05).
Table 3
Changes in endpoints before and after AVR
| Pre AVR ( n = 113) | Post AVR ( n = 113) | Change | p – value for change | |
|---|---|---|---|---|
| Functional parameters | ||||
| NYHA class III+ | 38.5 | 4.0 | −34.5 | <0.001 |
| 6-minute walk distance, m | 510 (360-600) | 570 (450-686) | +94 (+58 to +131) | <0.001 |
| Imaging parameters | ||||
| Mean gradient, mmHg | 47 (39-57) | 11 (8-16) | −35 (−37 to −32) | <0.001 |
| LVM index, g/m 2 | 87 (70-107) | 68 (57-84) | −19 (−21 to −16) | <0.001 |
| LVEF, % | 74 (64-81) | 77 (69-82) | +3 (+1 to +6) | 0.009 |
| LVEDV, mL | 118 (94-147) | 111 (91-141) | −8 (−15 to −2) | 0.017 |
| LVESV, mL | 29 (19-49) | 25 (17-39) | −10 (−16 to −4) | 0.002 |
| LV stroke volume, mL | 80 (67-95) | 81 (68-100) | +2 (−3 to +6) | 0.45 |
| T1 relaxation times precontract, ms | 1048 (1019-1073) | 1034 (1006-1053) | −7 (−16 to +3) | 0.17 |
| ECV, % | 28 (27-30) | 29 (28-31) | +2 (+1 to +2) | <0.001 |
| Laboratory results | ||||
| NT-pro BNP, pg/mL | 76 (29-177) | 39 (23-107) | −86 (−132 to −41) | <0.001 |
| Troponin, pg/mL | 14 (9-20) | 12 (9-18) | −4 (−7 to −1) | 0.005 |
| Metabolomic profiling | ||||
| Glucose metabolism | ||||
| Glucose, mmol/L | 4.9 (4.1-6.0) | 4.4 (4.0-5.0) | −0.7 (−1.0 to −0.3) | <0.001 |
| Lactate, mmol/L | 1.2 (1.0-1.5) | 1.2 (1.0-1.5) | +0.0 (−0.1 to +0.1) | 0.94 |
| Citrate, mmol/L | 0.19 (0.17-0.22) | 0.19 (0.17-0.22) | +0.00 (−0.01 to +0.01) | 0.54 |
| Fatty acids | ||||
| Total fatty acids, mmol/L | 10.9 (9.4-12.7) | 10.4 (8.9-12.4) | −0.3 (−0.7 to +0.2) | 0.32 |
| Estimated degree of unsaturation | 1.23 (1.20-1.27) | 1.22 (1.17-1.26) | −0.02 (−0.03 to −0.01) | 0.005 |
| Omega-3 fatty acids, mmol/L | 4.1 (3.3-5.1) | 3.9 (3.1-4.7) | −0.03 (−0.05 to −0.01) | 0.008 |
| 22:6, docosahexaenoic acid, mmol/L | 1.3 (1.0-1.7) | 1.2 (0.9-1.5) | −0.02 (−0.02 to −0.01) | <0.001 |
| Omega-6 fatty acids, mmol/L | 3.5 (2.9-4.0) | 3.2 (2.7-3.7) | −0.2 (−0.4 to −0.1) | 0.011 |
| 18:2, linoleic acid, mmol/L | 2.8 (2.2-3.3) | 2.4 (2.0-3.0) | −0.2 (−0.4 to −0.1) | 0.006 |
| Polyunsaturated fatty acids, mmol/L | 3.9 (3.3-4.5) | 3.5 (3.0-4.2) | −0.2 (−0.4 to −0.1) | 0.009 |
| Monounsaturated fatty acids; 16:1, 18:1, mmol/L | 2.9 (2.3-3.4) | 2.8 (2.3-3.3) | +0.0 (−0.1 to +0.2) | 0.91 |
| Saturated fatty acids, mmol/L | 4.2 (3.7-4.7) | 4.0 (3.5-4.8) | −0.0 (−0.2 to +0.2) | 0.80 |
| Amino acids | ||||
| Nonbranched chained AA | ||||
| Alanine, mmol/L | 0.40 (0.36-0.45) | 0.43 (0.37-0.48) | +0.02 (+0.00 to +0.03) | 0.032 |
| Glutamine, mmol/L | 0.46 (0.41-0.52) | 0.49 (0.44-0.56) | +0.03 (+0.01 to +0.05) | 0.002 |
| Histidine, mmol/L | 0.05 (0.05-0.06) | 0.05 (0.05-0.06) | −0.001 (−0.002 to +0.001) | 0.56 |
| Branched chained AA | ||||
| Isoleucine, mmol/L | 0.07 (0.05-0.08) | 0.06 (0.05-0.08) | +0.002 (−0.003 to +0.006) | 0.40 |
| Leucine, mmol/L | 0.08 (0.07-0.10) | 0.09 (0.07-0.10) | +0.003 (−0.002 to +0.007) | 0.30 |
| Valine, mmol/L | 0.16 (0.14-0.19) | 0.16 (0.14-0.20) | +0.006 (−0.002 to +0.015) | 0.11 |
| Aromatic AA | ||||
| Phenylalanine, mmol/L | 0.06 (0.06-0.07) | 0.06 (0.06-0.07) | +0.001 (−0.001 to +0.003) | 0.43 |
| Tyrosine, mmol/L | 0.059 (0.053-0.071) | 0.062 (0.054-0.073) | +0.003 (+0.000 to + 0.006) | 0.034 |
| Ketone bodies | ||||
| Acetate, mmol/L | 0.05 (0.05-0.06) | 0.06 (0.05-0.06) | +0.01 (−0.01 to +0.04) | 0.34 |
| 3-hydroxybutyrate, mmol/L | 0.11 (0.10-0.14) | 0.12 (0.10-0.14) | +0.00 (−0.02 to +0.02) | 0.94 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
