Abstract
Background
Hyperuricemia (HUA) induces inflammation and insulin resistance and is reportedly associated with left ventricular hypertrophy (LVH) and possibly with left ventricular diastolic dysfunction (LVDD).
Objectives
To investigate associations among HUA, inflammation, and insulin resistance with LVDD.
Methods
We enrolled patients with metabolic syndrome (MetS) between August 1, 2017, and December 31, 2017. All participants underwent fasting blood tests and transthoracic echocardiography. HUA was defined as an serum uric acid level ≥ 7 mg/dl in men or ≥ 6 mg/dl in women. MetS was defined as at least three of the following Taiwanese criteria: central obesity, prehypertension, fasting glucose impairment, hypertriglyceridemia, and lower values of high-density lipoprotein cholesterol. LVDD was defined according to contemporary guidelines.
Results
The study included 63 patients (60% male) with a mean age of 53 ± 14 years and body mass index (BMI) of 29.4 ± 4.0 kg/m2. Prevalence rates of HUA, LVH, LVDD were 40%, 18%, and 10%, respectively. Baseline characteristics were similar between the HUA and normouricemia groups, except that the HUA group had significantly higher serum high-sensitivity interleukin 6 and tumor necrosis factor-alpha (TNF-α) levels. LVDD occurred more frequently in the HUA group (20.0% vs. 2.6%, p = 0.032). HUA was associated with LVDD [crude odds ratio (OR): 9.25, 95% confidence interval (CI): 1.01-84.7, p = 0.049]. In multivariate analysis, the most relevant factor associated with LVDD was TNF-α after adjustments for age, male sex, and body mass index (adjusted OR for TNF-α: 4.1, 95% CI: 1.02-16.5, p = 0.047).
Conclusions
The association between HUA and LVDD partially reflected a low-grade inflammation due to elevated TNF-α rather than increased insulin resistance in MetS patients.
Keywords: Echocardiography, Hyperuricemia, Left ventricular diastolic dysfunction, Metabolic syndrome
INTRODUCTION
Hyperuricemia (HUA) is defined as serum uric acid (SUA) levels exceeding the average physiologic concentration of 6.8 mg/dl in humans, and exhibits inflammatory properties due to reactive oxidative stress, and therefore induces vascular inflammation, insulin resistance, and adipocyte dysfunction.1-3 HUA is reportedly associated with obesity, hypertension and metabolic syndrome (MetS),4-6 which are traditional risk factors for left ventricular hypertrophy (LVH) and left ventricular (LV) diastolic dysfunction (LVDD).4,7,8 Although HUA and MetS share common pathologic factors and are both risk factors for diabetes, HUA might induce LVDD by causing systemic inflammation rather than by inducing insulin resistance. Indirect evidence from a previous study showed that the association between MetS and LVDD was independent of insulin resistance in individuals undergoing health examinations,9 although a conflicting result was found in patients with diabetes.7 The discrepancies between these studies can be mainly explained by the fact that HUA and MetS were both independently associated with LVDD, and an interaction between HUA and MetS has also been found in patients with diastolic heart failure.10 Therefore, we conducted the present study to investigate the relationships among HUA, LVDD, inflammation and insulin resistance in patients with MetS.
METHODS
We prospectively enrolled patients with MetS aged between 20 and 75 years from a single hospital between August 1, 2017, and December 31, 2017. The exclusion criteria were as follows: (1) LV ejection fraction < 40%; (2) severe valvular heart disease; (3) significant structural heart disease or myocardial diseases, such as hypertrophic, dilated, infiltrative or restrictive cardiomyopathy, or congenital heart disease; (4) persistent or chronic atrial fibrillation; (5) history of cardiac surgery, including coronary artery bypass surgery or valve intervention; (6) history of intracardiac device implantation; (7) chronic obstructive pulmonary disease; (8) severe anemia; or (9) obvious systemic disease. We conducted the present study in accordance with the Declaration of Helsinki. All study participants signed informed consent and underwent body composition measurements, blood tests and echocardiography. Body height, weight and composition were measured using an OMRON HBF-701 device. Blood samples were obtained after the participants had fasted overnight. The first author performed transthoracic echocardiography according to contemporary guidelines with a Philips IE 33 system. The study patients underwent blood tests before echocardiography within an interval of less than one week. The study protocol was approved by the Tri-Service General Hospital Institutional Review Board (number TSGH 2-106-05-078). The study protocol was registered at ClinicalTrials.gov (no. NCT03495999).
Blood tests included measurement of electrolytes, lipid, glucose, glycosylated hemoglobin, insulin, alanine aminotransferase, creatinine, uric acid, high-sensitivity C-reactive protein (hs-CRP), high-sensitivity interleukin 6 (hs-IL-6), and tumor necrosis factor-alpha (TNF-α) levels. The homeostasis model assessment for insulin resistance (HOMA-IR) was calculated as "fasting plasma insulin (mIU/L) × fasting plasma glucose (mg/dl) / 405". Levels of hs-CRP were analyzed immediately after the blood samples had been collected using a latex-enhanced immunoturbidimetric assay (ADVIA 1800, ADVIA Chemistry XPT, Dimension RXL, SIEMENS, AU 640, Beckman Coulter) at a qualified laboratory. Whole blood was centrifuged at 3,000 rounds per minute for 15 minutes at room temperature, and then, plasma was stored at -80 °C. We used commercial enzyme-linked immunosorbent assay (ELISA) kits to measure hs-IL-6 (human IL-6 Quantikine HS ELISA Kit, R & D) and TNF-α (Human TNF-alpha Quantikine ELISA Kit, R & D) in a batch at the end of enrollment in the study.
According to the Taiwan criteria and our previous studies,4,5,11 patients with MetS must fulfill at least three of the five following criteria: (1) waist circumference > 90 cm in men or > 80 cm in women; (2) blood pressure > 130/85 mmHg or already on antihypertensive medication; (3) fasting glucose ≥ 100 mg/dl or already on antidiabetic agents; (4) triglyceride level ≥ 150 mg/dl or already on lipid-lowering therapy for hypertriglyceridemia; and (5) high-density lipoprotein cholesterol ≤ 40 mg/dl in men or ≤ 50 mg/dl in women.11 HUA was defined as an SUA level ≥ 7 mg/dl in men or ≥ 6 mg/dl in women, as widely used in clinical practice and in previous studies.2,4,5,12 Echocardiographic LVH was defined as an LV mass index > 115 g/m2 in men or > 95 g/m2 in women, according to contemporary guidelines.13,14 In patients with normal LV ejection fraction, the contemporary definition of LVDD involves > 50% positivity for the following criteria: (1) average E/e′ ≥ 14, (2) left atrial volume index (LAVI) > 34 ml/m2, (3) tricuspid regurgitation (TR) velocity > 2.8 m/s, and (4) septal E′ velocity < 7 cm/s or lateral E′ < 10 cm/s. Patients with < 50% positive criteria were classified as having normal diastolic function, and those with 50% positive criteria were classified as having indeterminate diastolic function. We then used the algorithm of mitral inflow to estimate LV filling pressures and grading LVDD in the patients with > 50% positive criteria according to the contemporary guidelines. Grade II LVDD was defined as > 50% positivity for the first three criteria (average E/e′ > 14, LAVI > 34 ml/m2, TR velocity > 2.8 m/s).13,14
Continuous variables are expressed as the mean and standard deviation (SD), and categorical variables are expressed as numbers and percentages. We used independent t tests to compare the means of continuous variables between the two groups, and Pearson’s chi-squared tests were used to determine whether the distributions of categorical variables were significantly different. The presence of grade II LVDD was the focus of our study, and we first used univariate logistic regression analysis to examine the associations between our study focus and each variable, including inflammatory biomarkers and insulin resistance. Confounders were adjusted for in multivariate logistic regression analysis. We also used scatter plots to illustrate the distributions of LVDD, SUA and TNF-α in subgroups classified according to an estimated glomerular filtration rate (eGFR) of 60 ml/min/1.73 m2. All p values were two-tailed, and considered significant if p < 0.05. We performed statistical analyses using R software, version 3.4.1 (The R Project for Statistical Computing).
RESULTS
We enrolled 67 candidates initially and excluded one patient with regional wall motion abnormalities of the myocardium diagnosed by echocardiography. Three patients did not receive blood tests. Ultimately, 63 patients with a mean age of 53 ± 14 years, a mean body mass index (BMI) of 29.4 ± 4.0 kg/m2, and a mean eGFR of 92 ± 24 ml/min/1.73 m2 were analyzed, of whom 60% were male. The HUA group (N = 25) and the normouricemic group (N = 38) had similar baseline characteristics, body composition, medication prescriptions and laboratory data, except that SUA, hs-IL-6 and TNF-α levels were significantly greater in the HUA group than in the normouricemic group (Table 1). The HUA group had greater hs-IL-6 and TNF-α levels than the control group (3.0 ± 1.8 pg/ml vs. 2.2 ± 1.3 pg/ml, p = 0.048 for hs-IL-6 and 2.7 ± 1.0 pg/ml vs. 2.3 ± 0.8 pg/ml, p = 0.042 for TNF-α); however, hs-CRP levels and insulin resistance were not significantly different between the two groups (Figure 1).
Table 1. Baseline characteristics, laboratory data, and medication use in patients with metabolic syndrome.
| Normouricemia (N = 38) | Hyperuricemia (N = 25) | p | |
| Age (years) | 52.3 ± 11.1 | 54.8 ± 17.6 | 0.395 |
| Male | 23 (61.0) | 15 (60.0) | 1.000 |
| Body mass index (kg/m2) | 29.4 ± 3.5 | 29.4 ± 4.7 | 0.929 |
| Waist circumference (cm) | 96.7 ± 8.4 | 100.1 ± 13.6 | 0.147 |
| Systolic blood pressure (mmHg) | 132 ± 17 | 134 ± 21 | 0.686 |
| Diastolic blood pressure (mmHg) | 80 ± 11 | 75 ± 11 | 0.092 |
| Smoking | 14 (36.8) | 6 (24) | 0.408 |
| Hypertension | 30 (78.9) | 15 (60) | 0.154 |
| Diabetes | 19 (50) | 11 (44) | 0.797 |
| Dyslipidemia | 35 (92.1) | 23 (92) | 1.000 |
| Laboratory data | |||
| Uric acid (mg/dl) | 5.4 ± 0.8 | 7.5 ± 0.8 | < 0.001 |
| Glucose (mg/dl) | 114 ± 24 | 127 ± 49 | 0.174 |
| Insulin (mIU/L) | 18.4 ± 20.9 | 16.6 ± 11.5 | 0.699 |
| HOMA-IR | 5.1 ± 5.9 | 5.2 ± 4.2 | 0.945 |
| Total cholesterol (mg/dl) | 168 ± 48 | 176 ± 39 | 0.396 |
| High-density lipoprotein-cholesterol (mg/dl) | 40 ± 10 | 39 ± 11 | 0.748 |
| Low-density lipoprotein-cholesterol (mg/dl) | 101 ± 30 | 105 ± 33 | 0.586 |
| Triglyceride (mg/dl) | 167 ± 309 | 199 ± 108 | 0.180 |
| Creatinine (mg/dl) | 0.9 ± 0.3 | 0.9 ± 0.3 | 0.637 |
| Creatinine clearance by Cockcroft-Gault equation (ml/min) | 113 ± 39 | 111 ± 55 | 0.901 |
| Estimated glomerular filtration rate by MDRD equation (ml/min/1.73 m2) | 93 ± 24 | 89 ± 24 | 0.543 |
| Normal kidney function or CKD stage I | 20 ± 52.6 | 11 ± 44 | 0.752 |
| CKD stage II | 15 ± 39.5 | 11 ± 44 | |
| CKD stage III | 3 ± 7.9 | 3 ± 12 | |
| Alanine aminotransferase (U/L) | 34 ± 30 | 33 ± 29 | 0.861 |
| Hemoglobin (g/dl) | 14.0 ± 2.8 | 13.7 ± 4.4 | 0.713 |
| Medication use | |||
| Anti-hypertensive agents | 30 (78.9) | 15 (60) | 0.154 |
| The number of anti-hypertensive agents* | 2.3 ± 1.3 | 2.1 ± 1.2 | 0.677 |
| Beta-blocker | 12 (31.6) | 4 (16) | 0.239 |
| Calcium channel blocker | 20 (52.6) | 11 (44) | 0.609 |
| ACEI or ARB | 25 (65.8) | 13 (52) | 0.303 |
| Thiazide diuretics | 6 (15.8) | 2 (8) | 0.461 |
| Spironolactone | 2 (5.3) | 0 (0) | 0.514 |
| Anti-diabetic agents | 18 (47.4) | 11 (44) | 1.000 |
| The number of anti-diabetic agents* | 2.4 ± 1.0 | 2.6 ± 0.8 | 0.606 |
| Metformin | 17 (44.7) | 9 (36) | 0.603 |
| Sulfonylurea | 9 (23.7) | 7 (28) | 0.772 |
| Thiazolidinediones | 3 (7.9) | 0 (0) | 0.270 |
| Dipeptidyl peptidase IV inhibitor | 6 (15.8) | 7 (28) | 0.341 |
| Sodium–glucose co-transporter type 2 inhibitor | 4 (10.5) | 2 (8) | 1.000 |
| Carbohydrase inhibitor | 2 (5.3) | 2 (8) | 1.000 |
| Insulin | 3 (7.9) | 2 (8) | 1.000 |
| Lipid lowering therapy | 22 (57.9) | 12 (48) | 0.606 |
| Urate lowering therapy | 3 (7.9) | 1 (4) | 1.000 |
Values were expressed as mean ± standard deviation or number (percentage).
ACEI, angiotensin-converting enzyme inhibitors; ARB, angiotensin receptor blockers; CKD, chronic kidney disease; HOMA-IR, homeostatic model assessment for insulin resistance; MDRD, Modification of Diet in Renal Disease study.
* The mean number of anti-hypertensive and anti-diabetic agents were calculated in the 45 and 29 patients with the use of the medication, not including the patients without the use of the medication.
Figure 1.
Patients in the HUA group had significantly greater serum concentrations of systemic inflammatory biomarkers, such as TNF-α and hs-IL-6, than did the normouricemic group, but insulin resistance was not significantly different between the two groups. HOMA-IR, homeostasis model assessment for insulin resistance; hs-CRP, high-sensitivity C-reactive protein; hs-IL-6, high-sensitivity interleukin 6; HUA, hyperuricemia; TNF-α, tumor necrosis factor-alpha.
The HUA group had lower values of interventricular septum thickness diameter (10.2 ± 1.1 mm vs. 10.9 ± 1.6 mm, p = 0.037) and LV mass index (85 g/m2 vs. 96 g/m2, p = 0.049) and a lower prevalence of echocardiographic LVH (12% vs. 21%, p = 0.503) than the control group (shown in Table 2). Of the 63 patients, six had diastolic dysfunction and 18 had indeterminate diastolic function, and the other 39 patients had normal diastolic dysfunction. The six patients with diastolic dysfunction and the two positive criteria of mitral inflow were further classified as grade II LVDD. The ratio of grade II LVDD was significantly higher in the HUA group than in the normouricemic group (20.0% vs. 2.6%, p = 0.032). A study flow chart describing the process of classifying the patients into normal diastolic function, indeterminate LV diastolic function, and LVDD is shown in Figure 2.
Table 2. Echocardiographic parameters in patients with metabolic syndrome.
| Normouricemia (N = 38) | Hyperuricemia (N = 25) | p | |
| Aorta diameter (mm) | 31.0 ± 3.1 | 29.4 ± 2.9 | 0.055 |
| LA (mm) | 39.3 ± 4.6 | 39.0 ± 4.4 | 0.804 |
| Interventricular septum thickness diameter (mm) | 10.9 ± 1.6 | 10.2 ± 1.1 | 0.037 |
| Posterior wall thickness diameter (mm) | 10.4 ± 1.8 | 10.5 ± 1.5 | 0.665 |
| LV diameter in end-diastole (mm) | 47.5 ± 4.3 | 45.5 ± 5.1 | 0.099 |
| LV diameter in end-systole (mm) | 28.8 ± 3.2 | 28.0 ± 4.7 | 0.442 |
| LV ejection fraction (%) | 69.3 ± 7.5 | 69.0 ± 6.0 | 0.848 |
| LV mass (g) | 185.4 ± 50.1 | 168.4 ± 44.1 | 0.171 |
| LV mass index (g/m2) | 95.9 ± 22.9 | 85.2 ± 16.3 | 0.049 |
| LV hypertrophy defined by LV mass index | 8 (21.1) | 3 (12.0) | 0.503 |
| Aortic velocity (cm/s) | 124.9 ± 25.2 | 130.3 ± 23.4 | 0.402 |
| TR velocity (cm/s)* | 219.8 ± 28.5 | 202.2 ± 46.9 | 0.207 |
| TR pressure gradient (mmHg)* | 19.7 ± 4.9 | 18.6 ± 8.7 | 0.650 |
| TR velocity > 280 cm/s | 0 (0) | 0 (0) | - |
| Mitral E-wave velocity (cm/s) | 74.1 ± 22.1 | 78.7 ± 17.7 | 0.391 |
| Mitral E-wave velocity ≤ 50 cm/s | 4 (10.5) | 0 (0) | 0.145 |
| Mitral A-wave velocity (cm/s) | 82.1 ± 19.4 | 85.9 ± 21.7 | 0.476 |
| Mitral E-velocity deceleration time (ms) | 189.0 ± 48.1 | 218.4 ± 62.1 | 0.038 |
| Mitral E/A ratio | 0.9 ± 0.3 | 1.0 ± 0.3 | 0.699 |
| Mitral E/A ratio ≤ 0.8 | 16 (42.1) | 9 (36.0) | 0.793 |
| Mitral E/A ratio between 0.8 and 2 | 22 (65.8) | 16 (64) | 0.245 |
| Mitral E/A ratio ≥ 2 | 0 (0) | 0 (0) | |
| Mitral inflow and E/A ratio categories | |||
| Mitral E/A ratio ≤ 0.8 and E velocity ≥ 50 cm/s | 12 (31.6) | 9 (36) | |
| Mitral E/A ratio ≤ 0.8 and E velocity ≤ 50 cm/s | 4 (10.5) | 0 (0) | |
| Septal e′ velocity (cm/s) | 6.8 ± 2.3 | 6.9 ± 2.9 | 0.960 |
| Septal e′ velocity < 7 cm/s | 25 (65.8) | 14 (56) | 0.596 |
| Septal a′ velocity (cm/s) | 8.6 ± 1.4 | 8.3 ± 1.8 | 0.458 |
| Septal E/e′ | 11.3 ± 3.4 | 13.6 ± 4.6 | 0.037 |
| Lateral e′ velocity (cm/s) | 8.5 ± 3.4 | 8.9 ± 3.7 | 0.655 |
| Lateral e′ velocity < 10 cm/s | 23 (60.5) | 15 (60.0) | 1.000 |
| Lateral a′ velocity (cm/s) | 10.6 ± 3.0 | 9.8 ± 3.4 | 0.316 |
| Lateral E/e′ | 9.0 ± 5.3 | 9.1 ± 4.7 | 0.897 |
| Septal e′ velocity < 7 cm/s or lateral e′ velocity < 10 cm/s | 30 (78.9) | 18 (72.0) | 0.558 |
| LAVI (A4C) (mL/m2) | 26.1 ± 9.1 | 28.5 ± 13.3 | 0.391 |
| LAVI (A2C) (mL/m2) | 27.3 ± 9.5 | 26.0 ± 11.5 | 0.626 |
| LAVI (BP) (mL/m2) | 28.3 ± 7.6 | 28.0 ± 8.4 | 0.846 |
| LAVI (BP) > 34 ml/m2 | 10 (26.3) | 6 (24.0) | 1.000 |
| Average e′ velocity (cm/s) | 7.7 ± 2.4 | 7.9 ± 3.1 | 0.750 |
| Average a′ velocity (cm/s) | 9.6 ± 1.9 | 9.1 ± 2.4 | 0.302 |
| Average E/e′ | 10.1 ± 3.9 | 11.3 ± 4.2 | 0.233 |
| Average E/e′ > 14 | 7 (18.4) | 9 (36.0) | 0.145 |
| The number of LV diastolic function criteria | 0.063 | ||
| 0 | 7 (18.4) | 6 (24.0) | |
| 1 | 16 (42.1) | 10 (40.0) | |
| 2 | 14 (36.8) | 4 (16.0) | |
| 3 | 1 (2.6) | 5 (20.0) | |
| 4 | 0 (0) | 0 (0) | |
| Normal diastolic function | 23 (60.5) | 16 (64.0) | 1.000 |
| Indeterminate LV diastolic function | 14 (36.8) | 4 (16.0) | 0.092 |
| LVDD | 1 (21.1) | 5 (24.0) | 1.000 |
| The number of mitral inflow criteria | 0.037 | ||
| 0 | 22 (57.9) | 15 (60.0) | |
| 1 | 15 (39.5) | 5 (20.0) | |
| 2 | 1 (2.6) | 5 (20.0) | |
| Grade I LVDD | 0 (0) | 0 (0) | - |
| Grade II LVDD | 1 (2.6) | 5 (20.0) | 0.032 |
Values were expressed as mean ± standard deviation or number (percentage).
A2C, apical two chamber view; A4C, apical four chamber view; BP, biplane; LA, left atrium; LAP, left atrial pressure; LAVI, LA maximum volume index; LVDD, left ventricular diastolic dysfunction; TR, tricuspid regurgitation.
* TR velocity and pressure gradient were reported in the 31 patients given that the parameters cannot be detected in the rest of 32 patients.
Figure 2.
The diastolic dysfunction diagnosis algorithm based on 2016 ASE/EACVI diastolic function guidelines. ASE, American Society of Echocardiography; EACVI, European Association of Cardiovascular Imaging; HUA, hyperuricemia; LA, left atrium; LV EF, left ventricular ejection fraction; NUA, normouricemia; TR, tricuspid regurgitation.
In univariate logistic regression analysis, HUA was significantly associated with grade II LVDD [odds ratio (OR): 9.25, 95% confidence interval (CI): 1.01-84.7, p = 0.049], although HUA was not associated with indeterminate diastolic function (OR: 0.327, 95% CI: 0.093-1.147, p = 0.081). Other variables associated with grade II LVDD included age (OR: 1.222, 95% CI: 1.054-1.416, p = 0.008), male sex (OR: 0.108, 95% CI: 0.012-0.990, p = 0.049), and TNF-α (OR: 2.616, 95% CI: 1.165-5.872, p = 0.020), but not hs-CRP (OR: 2.727, 95% CI: 0.507-14.656, p = 0.242) or hs-IL-6 levels (OR: 1.066, 95% CI: 0.640-1.774, p = 0.807). After adjusting for age, male sex, and BMI, HUA was not associated with grade II LVDD in the multivariate logistic regression analysis (adjusted OR: 1.515, 95% CI: 0.052-44.58, p = 0.81), however TNF-α was still associated with grade II LVDD (adjusted OR: 4.100, 95% CI: 1.018-16.52, p = 0.047).
We further divided the study patients into subgroups with an eGFR of more or less than 60 ml/min/1.73 m2. All patients with grade II LVDD had HUA, except that the only patient treated with urate-lowering therapy had a normal SUA level. All patients with TNF-α levels greater than the mean of 2.4 pg/ml had grade II LVDD (Figure 3).
Figure 3.
TNF-α, rather than SUA, was significantly associated with grade II LVDD in subgroups stratified by eGFR. The vertical and horizontal dotted lines represent the mean TNF-α (2.4 pg/ml) and uric acid (6.2 mg/dl) levels, respectively. eGFR, estimated glomerular filtration rate; LAP, left atrial pressure; LVDD, left ventricular diastolic dysfunction; SUA, serum uric acid; TNF-α, tumor necrosis factor-alpha.
In order to evaluate the impact of indeterminate diastolic function on the study results, we re-classified the 18 patients with indeterminate diastolic function by the presence of echocardiographic LVH; the 12 patients with normal values of LV mass index were re-classified as having normal diastolic function, and the other six patients with echocardiographic LVH were re-classified as having LVDD. The six patients had one positive criteria of mitral inflow, and their grade of LVDD was re-classified as grade I. Although the re-classification process determined that the six patients had grade I LVDD, the association between SUA and grade II LVDD did not change.
DISCUSSION
HUA was significantly associated with the presence of grade II LVDD in the present investigation. A previous study reported that insulin resistance was associated with LVDD in patients with MetS compared to those without MetS,9 and another study similarly reported that subjects with concomitant HUA and MetS had a higher prevalence of echocardiographic LVH than subjects with normouricemia but without MetS.15 We consistently showed that HUA could be used as a predictor for LVDD in patients with insulin resistance. We also showed that the HUA patients had lower values of LV mass index and a lower prevalence of echocardiographic LVH than the normouricemic patients, however this result seems to conflict with those of other studies.4,16 We previously showed a dose-response effect of elevated SUA levels on the prevalence of electrocardiographic LVH in healthy individuals,4 and another case-control study showed that the highest vs. lowest quartile was associated with the incidence of echocardiographic LVH.16 These major discrepancies can mainly be explained by the diversity of the study population and the level of kidney function. The present study enrolled MetS patients, and half of them had chronic kidney disease, whereas approximately 80% of the healthy individuals in our previous study had normal kidney function;4 kidney function was not reported in the other study.16 Given that LVH induces LVDD,13 the lower prevalence of echocardiographic LVH in our patients with HUA might indirectly imply that the effect of HUA might be more pronounced depending on the grade of LVDD in the present study, and this assumption may be supported by a previous study.17 Lin et al. showed that HUA patients had greater LV mass index values and a greater prevalence of moderate-to-severe LVDD than normouricemic subjects. Our study results seem to partially conflict with the study by Lin et al.17 who showed that gout, rather than HUA alone, was associated with LVDD. This inconsistency may be mainly explained by marked differences in study populations between the two studies; patients with moderate insulin resistance were enrolled in the present study, whereas few patients had diabetes in the former study. The prevalence rates of hypertension and dyslipidemia were also much greater in our study than in the study by Lin et al.17 This implies that our patients, even those without gout, had an increased inflammatory status compared to the patients in the study by Lin et al.,17 and this may explain why HUA was associated with LVDD in the univariate logistic regression analysis in our study.
We further showed that after HUA, TNF-α was associated with LVDD in MetS patients in multivariate logistic regression analysis. Because the significant association between HUA and LVDD became nonsignificant after adjusting for TNF-α, HUA might partially serve as a surrogate biomarker for TNF-α, which is associated with the presence of grade II LVDD in MetS patients. Although the independent association between SUA levels and LVDD was not demonstrated in our study, we still found that TNF-α, which is associated with SUA, was an independent predictor of LVDD and provided complimentary evidence for the association between SUA and LVDD, as shown in previous studies.18-20 In our study, the hyperuricemic and normouricemic patients had similar baseline characteristics and laboratory indices; therefore, elevated circulating levels of TNF-α in HUA vs. normouricemic patients may also be induced by elevated SUA levels and not only by comorbidities such as hypertension, diabetes and obesity, as described in a previous review.21 Furthermore, elevated TNF-α levels have also been found in patients with heart failure with preserved ejection compared with healthy individuals,22 and TNF-α levels are associated with incident heart failure in the elderly.23 Although we did not recruit healthy individuals as control subjects, we assume that TNF-α can be used to predict LVDD preceding incident heart failure in MetS patients and the general population.
We suggest urate-lowering therapies to reduce systemic inflammation in patients with concomitant MetS and HUA rather than directly initiating anti-TNF-α therapy, although circulating serum TNF-α levels have been associated with heart failure. TNF-α inhibitors are known to be associated with incident heart failure and heart failure exacerbation,24 and the systemic side effects of TNF-α inhibitors can overwhelm the benefits of decreasing systemic inflammation or improving LVDD.25 Regarding the safety and cost-effectiveness of treatment, urate-lowering therapies including uricosuric agents or xanthine oxidase inhibitors might be reasonable alternates to improve LVDD, as these treatments can lower both SUA and TNF-α levels.26 A randomized controlled trial investigated the effect of a uricosuric agent on improvements in chronic heart failure, and showed that patients randomized to benzbromarone vs. placebo did not exhibit hemodynamic improvements of heart function, including left ventricular ejection fraction and brain natriuretic peptide. The trial concluded that lowering SUA without inhibiting xanthine oxidase was not associated with an improvement in chronic heart failure.26 Another randomized controlled trial with a study period of 12 months showed that xanthine oxidase inhibitors allopurinol and febuxostat may improve cardiac function in patients with chronic heart failure and hyperuricemia by reducing inflammation.27 Systematic reviews and meta-analyses have shown that urate-lowering therapies are not associated with the incidence of cardiovascular events,28,29 whereas a counterargument was shown in the CARES trial, with overwhelming numbers of patients with high cardiovascular risk factors randomized to treatment with febuxostat vs. allopurinol.30 Since the costs of time and budget are enormous in any large trial investigating the effects of urate-lowering therapies on cardiovascular outcomes, we conducted an open-label randomized controlled trial to investigate the effects of urate-lowering therapies on changes in LVDD and surrogate biomarkers (NCT03534037 at ClinicalTrials.gov).
According to contemporary guidelines,13,14 these is no definition for normal ejection fraction with myocardial disease, and no standard process exists to evaluate indeterminate diastolic function. Therefore, we tried to identify a cut-off value of global longitudinal strain (GLS) to define LVDD in patients with MetS, and investigated the association between GLS and LVDD. In patients with indeterminate diastolic function, determining the presence of LVDD by GLS is rather difficult, because there is currently no consensus on the ideal cut-off value of GLS. GLS values have been reported to range widely from -8% in patients with systolic heart failure to -15% in patients with symptomatic severe aortic stenosis.31,32 In a review article, the values of GLS varied widely in various study populations,33 and little evidence discussed the association between GLS and LVDD in "patients with MetS". Systolic strain alteration may exist despite normal diastolic function in patients with diabetes,34 and GLS can be abnormal in the absence of LVDD.35 In other words, a decreased GLS can proceed the presence of LVDD.
The present study has several limitations, including the small number of patients and lack of healthy control subjects. We cannot precisely explain why a lower LV mass index was found in the patients with hyperuricemia vs. normouricemia in our study, but the limited number of cases may be a factor. Although the HUA group had a greater prevalence of grade II LVDD than the normouricemic group, type I errors could have occurred with respect to the limited number of subjects and the number of indeterminate LVDD. We statistically adjusted only for age, male sex and BMI in multivariate logistic regression analysis, given that the baseline characteristics and laboratory data were not significantly different, but the association was nonsignificant in the full-model adjustment due to the limited number of subjects. We did not measure natriuretic peptides that are well acknowledged to be powerful biomarkers of heart failure, but an important trial has already shown that LVDD may be used to predict cardiac outcomes independently of laboratory indices.36 To resolve these limitations, another prospective cohort study conducted by our study group is ongoing to evaluate the association between HUA and LVDD in MetS patients and healthy individuals. We aim to elucidate the relationship between SUA and other biomarkers that are indicators of systematic inflammation, cardiac fibrosis, and atherosclerosis in the future.
CONCLUSIONS
The association between HUA and grade II LVDD partially reflected low-grade inflammation due to elevated TNF-α rather than increased insulin resistance in MetS patients. Future randomized controlled trials may investigate the effect of urate-lowering therapies on improvements in low-grade inflammation and LVDD in HUA patients with MetS.
Acknowledgments
The study was partly supported by academic research grants from the Taiwan Society of Cardiology (project number: 1070506) and Tri-Service General Hospital Songshan Branch (project number: 10811). The authors designed the study, collected the data, performed statistical analyses, wrote the report and made the decision to submit the article for publication.
CONFLICT OF INTEREST
All the authors declare no conflict of interest.
FUNDING
The study was supported by Taiwan Society of Cardiology. The authors designed the study, collected the data, performed statistical analyses, wrote the report and made the decision to submit the article for publication.
REFERENCES
- 1.Feig DI, Kang DH, Johnson RJ. Uric acid and cardiovascular risk. N Engl J Med. 2008;359:1811–1821. doi: 10.1056/NEJMra0800885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Maiuolo J, Oppedisano F, Gratteri S, et al. Regulation of uric acid metabolism and excretion. Int J Cardiol. 2016;213:8–14. doi: 10.1016/j.ijcard.2015.08.109. [DOI] [PubMed] [Google Scholar]
- 3.Tsushima Y, Nishizawa H, Tochino Y, et al. Uric acid secretion from adipose tissue and its increase in obesity. J Biol Chem. 2013;288:27138–27149. doi: 10.1074/jbc.M113.485094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Liu CW, Chen KH, Tseng CK, et al. The dose-response effects of uric acid on the prevalence of metabolic syndrome and electrocardiographic left ventricular hypertrophy in healthy individuals. Nutr Metab Cardiovasc Dis. 2018;29:30–38. doi: 10.1016/j.numecd.2018.10.001. [DOI] [PubMed] [Google Scholar]
- 5.Liu CW, Chang WC, Lee CC, et al. Hyperuricemia is associated with a higher prevalence of metabolic syndrome in military individuals. Mil Med. 2018;183:e391–e395. doi: 10.1093/milmed/usy097. [DOI] [PubMed] [Google Scholar]
- 6.Zaragoza-García O, Navarro-Zarza JE, Maldonado-Anicacio JY, et al. Hypertriglyceridemic waist is associated with hyperuricemia and metabolic syndrome in rheumatoid arthritis patients. Diabetes Metab Syndr. 2019;13:722–729. doi: 10.1016/j.dsx.2018.11.053. [DOI] [PubMed] [Google Scholar]
- 7.Bajraktari G, Koltai MS, Ademaj F, et al. Relationship between insulin resistance and left ventricular diastolic dysfunction in patients with impaired glucose tolerance and type 2 diabetes. Int J Cardiol. 2006;110:206–211. doi: 10.1016/j.ijcard.2005.08.033. [DOI] [PubMed] [Google Scholar]
- 8.Wang J, Qin T, Chen J, et al. Hyperuricemia and risk of incident hypertension: a systematic review and meta-analysis of observational studies. PLoS One. 2014;9:e114259. doi: 10.1371/journal.pone.0114259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hwang YC, Jee JH, Kang M, et al. Metabolic syndrome and insulin resistance are associated with abnormal left ventricular diastolic function and structure independent of blood pressure and fasting plasma glucose level. Int J Cardiol. 2012;159:107–111. doi: 10.1016/j.ijcard.2011.02.039. [DOI] [PubMed] [Google Scholar]
- 10.Tang ZH, Fang Z, Zeng F, et al. Association and interaction analysis of metabolic syndrome and serum uric acid on diastolic heart failure. J Endocrinol Invest. 2013;36:579–583. doi: 10.3275/8851. [DOI] [PubMed] [Google Scholar]
- 11.Hwang LC, Bai CH, Chen CJ. Prevalence of obesity and metabolic syndrome in Taiwan. J Formos Med Assoc. 2006;105:626–635. doi: 10.1016/S0929-6646(09)60161-3. [DOI] [PubMed] [Google Scholar]
- 12.Liu CW, Liao PC, Chen KC, et al. Relationship of serum uric acid and killip class on mortality after acute ST-segment elevation myocardial infarction and primary percutaneous coronary intervention. Int J Cardiol. 2017;226:26–33. doi: 10.1016/j.ijcard.2016.10.025. [DOI] [PubMed] [Google Scholar]
- 13.Nagueh SF, Smiseth OA, Appleton CP, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2016;17:1321–1360. doi: 10.1093/ehjci/jew082. [DOI] [PubMed] [Google Scholar]
- 14.Nagueh SF, Smiseth OA, Appleton CP, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2016;29:277–314. doi: 10.1016/j.echo.2016.01.011. [DOI] [PubMed] [Google Scholar]
- 15.Yu S, Guo X, Yang H, Sun Y. Combination of hyperuricemia and metabolic syndrome is an independent and powerful predictor for left ventricular hypertrophy in rural Chinese. Ann Endocrinol. 2015;76:264–271. doi: 10.1016/j.ando.2015.01.002. [DOI] [PubMed] [Google Scholar]
- 16.Cuspidi C, Facchetti R, Bombelli M, et al. Uric acid and new onset left ventricular hypertrophy: findings from the PAMELA population. Am J Hypertens. 2017;30:279–285. doi: 10.1093/ajh/hpw159. [DOI] [PubMed] [Google Scholar]
- 17.Lin JC, Lin CL, Chen MC, et al. Gout, not hyperuricemia alone, impairs left ventricular diastolic function. Arthritis Res Ther. 2015;17:323. doi: 10.1186/s13075-015-0842-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gromadzinski L, Januszko-Giergielewicz B, Pruszczyk P. Hyperuricemia is an independent predictive factor for left ventricular diastolic dysfunction in patients with chronic kidney disease. Adv Clin Exp Med. 2015;24:47–54. doi: 10.17219/acem/38159. [DOI] [PubMed] [Google Scholar]
- 19.Krishnan E, Hariri A, Dabbous O, Pandya BJ. Hyperuricemia and the echocardiographic measures of myocardial dysfunction. Congest Heart Fail. 2012;18:138–143. doi: 10.1111/j.1751-7133.2011.00259.x. [DOI] [PubMed] [Google Scholar]
- 20.Mocan M, Anton F, Suciu S, et al. Multimarker assessment of diastolic dysfunction in metabolic syndrome patients. Metab Syndr Relat Disord. 2017;15:507–514. doi: 10.1089/met.2017.0060. [DOI] [PubMed] [Google Scholar]
- 21.Paulus WJ, Tschöpe C. A novel paradigm for heart failure with preserved ejection fraction. J Am Col Cardiol. 2013;62:263. doi: 10.1016/j.jacc.2013.02.092. [DOI] [PubMed] [Google Scholar]
- 22.Putko BN, Wang Z, Lo J, et al. Circulating levels of tumor necrosis factor-alpha receptor 2 are increased in heart failure with preserved ejection fraction relative to heart failure with reduced ejection fraction: evidence for a divergence in pathophysiology. PLoS One. 2014;9:e99495. doi: 10.1371/journal.pone.0099495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kalogeropoulos A, Georgiopoulou V, Psaty BM, et al. Inflammatory markers and incident heart failure risk in older adults. J Am Col Cardiol. 2010;55:2129. doi: 10.1016/j.jacc.2009.12.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Javed Q, Murtaza I. Therapeutic potential of tumour necrosis factor-alpha antagonists in patients with chronic heart failure. Heart Lung Circ. 2013;22:323–327. doi: 10.1016/j.hlc.2012.12.002. [DOI] [PubMed] [Google Scholar]
- 25.Dogra S, Khullar G. Tumor necrosis factor-alpha antagonists: side effects and their management. Indian J Dermatol Venereol Leprol. 2013;79 Suppl 7:S35–S46. doi: 10.4103/0378-6323.115526. [DOI] [PubMed] [Google Scholar]
- 26.Ogino K, Kato M, Furuse Y, et al. Uric acid-lowering treatment with benzbromarone in patients with heart failure: a double-blind placebo-controlled crossover preliminary study. Circ Heart Fail. 2010;3:73–81. doi: 10.1161/CIRCHEARTFAILURE.109.868604. [DOI] [PubMed] [Google Scholar]
- 27.Nakagomi A, Saiki Y, Noma S, et al. Effects of febuxostat and allopurinol on the inflammation and cardiac function in chronic heart failure patients with hyperuricemia. IJC Metab Endocr. 2015;8:46–55. [Google Scholar]
- 28.Li S, Yang H, Guo Y, et al. Comparative efficacy and safety of urate-lowering therapy for the treatment of hyperuricemia: a systematic review and network meta-analysis. Sci Rep. 2016;6:33082. doi: 10.1038/srep33082. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Liu CW, Chang WC, Lee CC, et al. The net clinical benefits of febuxostat versus allopurinol in patients with gout or asymptomatic hyperuricemia - a systematic review and meta-analysis. Nutr Metab Cardiovasc Dis. 2019;29:1011–1022. doi: 10.1016/j.numecd.2019.06.016. [DOI] [PubMed] [Google Scholar]
- 30.White WB, Saag KG, Becker MA, et al. Cardiovascular safety of febuxostat or allopurinol in patients with gout. N Engl J Med. 2018;378:1200–1210. doi: 10.1056/NEJMoa1710895. [DOI] [PubMed] [Google Scholar]
- 31.Chan YH, Kuo CT, Wu LS, et al. Combined global longitudinal strain and intraventricular mechanical dyssynchrony predicts long-term outcome in patients with systolic heart failure. Circ J. 2016;80:177–185. doi: 10.1253/circj.CJ-15-0953. [DOI] [PubMed] [Google Scholar]
- 32.Lee HF, Hsu LA, Chan YH, et al. Prognostic value of global left ventricular strain for conservatively treated patients with symptomatic aortic stenosis. J Cardiol. 2013;62:301–306. doi: 10.1016/j.jjcc.2013.05.001. [DOI] [PubMed] [Google Scholar]
- 33.Potter E, Marwick TH. Assessment of left ventricular function by echocardiography: the case for routinely adding global longitudinal strain to ejection fraction. JACC Cardiovasc Imaging. 2018;11:260–274. doi: 10.1016/j.jcmg.2017.11.017. [DOI] [PubMed] [Google Scholar]
- 34.Ernande L, Bergerot C, Rietzschel ER, et al. Diastolic dysfunction in patients with type 2 diabetes mellitus: is it really the first marker of diabetic cardiomyopathy? J Am Soc Echocardiogr. 2011;24:1268–1275. doi: 10.1016/j.echo.2011.07.017. [DOI] [PubMed] [Google Scholar]
- 35.Yu B, Lin I, Cornish B, et al. Relationship between global longitudinal strain and left ventricular diastolic function. JACC. 2019;73:1675. [Google Scholar]
- 36.Shah AM, Claggett B, Sweitzer NK, et al. Cardiac structure and function and prognosis in heart failure with preserved ejection fraction: findings from the echocardiographic study of the treatment of preserved cardiac function heart failure with an aldosterone antagonist (TOPCAT) trial. Circ Heart Fail. 2014;7:740–751. doi: 10.1161/CIRCHEARTFAILURE.114.001583. [DOI] [PMC free article] [PubMed] [Google Scholar]