Gender differences in asymptomatic hyperuricemia’s effects on left ventricular function and remodeling in hypertension: insights from cardiac MRI

Abstract

Background

This study aimed to explore the association between asymptomatic hyperuricemia (HU) and left ventricular (LV) function and structure in patients with essential hypertension, using cardiac magnetic resonance (CMR). Additionally, it investigated gender differences in the impact of serum uric acid (SUA) on LV myocardial strain and remodeling.

Methods

A total of 244 essential hypertension patients (137 males and 107 females), comprising 98 with asymptomatic hyperuricemia [hypertension (HU+) group] and 146 without hyperuricemia [hypertension (HU-) group], and 62 controls were included in this study. Conventional LV function parameters and global LV myocardial strain parameters, namely, LV radial peak strain (GRPS), circumferential peak strain (GCPS), longitudinal peak strain (GLPS), were measured using CMR and compared among the three groups. Pearson correlation and regression analyses were conducted to assess the relationship between the physiological-biochemical index and CMR parameters in hypertensive patients of different genders.

Results

The LV GLPS progressively deteriorated from the control group to the hypertension (HU-) group, and further to the hypertension (HU+) group. LV GCPS and LV GRPS were reduced in the hypertension (HU+) group compared to the control and hypertension (HU-) groups. LV mass index (LVMI) and LV remodeling index increased gradually from the control group to the hypertension (HU-) group and then to the hypertension (HU+) group. SUA was significantly correlated with LV global peak strain. Multiple regression analyses showed SUA was independently associated with LV GLPS in both genders (β = -0.263, P = 0.025 in females and β = -0.328, P = 0.001 in males). After adjusting for clinical factors, SUA was independently associated with LVMI in male hypertensive patients (β = 0.189, P = 0.03), but not in females.

Conclusions

In hypertensive patients, elevated SUA is associated with impaired LV strain in both sexes; however, only in males is it independently related to greater LV mass index, indicating a gender‑specific influence on ventricular remodeling.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12872-026-05727-7.

Background

Hypertension is the most prevalent cardiovascular disease worldwide, with a continuously increasing burden. In clinical practice, different complications and concomitant diseases often aggravate the symptoms and prognosis of patients with hypertension [1, 2]. Due to interrelated metabolic factors, many patients with hypertension have elevated serum uric acid (SUA) levels. Yet, the absence of overt symptoms like gout or renal failure in the initial stages often leads to an underestimation of its potential clinical significance [3, 4]. Previous studies have established SUA as an independent risk factor for hypertension and cardiovascular events [5, 6]. However, the impact of HU on cardiac function in hypertensive patients, remains underexplored. Furthermore, sex hormones influence both SUA handling and myocardial remodeling, and the potential gender-specific differences in the effects of SUA on cardiac remodeling warrant further investigation [7–11].

Cardiac magnetic resonance imaging (CMR) is a non-invasive imaging modality that provides high-resolution, multi-parametric assessments of cardiac function and structure. Among CMR techniques, feature-tracking (FT) technology has emerged as a powerful tool for evaluating myocardial strain and predicting the prognosis of cardiovascular diseases [12]. Most prior investigations of hyperuricemia and myocardial deformation in hypertension have used speckle‑tracking echocardiography [13]; to our knowledge, no study has systematically evaluated left ventricular (LV) strain in this setting with feature‑tracking cardiac MRI. Compared with echocardiography, CMR offers superior spatial resolution, three‑dimensional coverage and excellent inter‑scanner reproducibility for global strain assessment [14].

Therefore, this study aimed to investigate the relationship between HU and LV dysfunction in patients with essential hypertension by CMR image. Additionally, we attempted to analyze whether increased SUA are related to LV alterations in hypertensive patients, with consideration for potential gender differences.

Materials and methods

Study population

This retrospective study was approved by the Ethics Committee on Biomedical Research, West China Hospital of Sichuan University (Approval No. 2019 − 811); the requirement for written informed consent was waived in accordance with the Chinese Measures for Ethical Review of Biomedical Research Involving Humans because all data were anonymised. This study was conducted in accordance with the Declaration of Helsinki. The study subjects were essential hypertensive patients who underwent CMR examination in our hospital from January 2012 to July 2024. Baseline demographic and clinical characteristics were obtained from clinical records for all participants. Hypertension was defined as systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg or being treated with antihypertensive drugs. Blood pressure values were obtained from our hospital’s electronic medical record system and correspond to the mean of two brachial measurements taken by trained nursing staff using a calibrated sphygmomanometer after the patient had been seated for at least five minutes. The exclusion criteria were heart failure, various congenital heart diseases, cardiomyopathy, coronary heart disease, severe valvular disease, atrial fibrillation, diabetes mellitus, severe hepatopulmonary dysfunction, secondary hypertension, body mass index (BMI) ≥ 30 and estimated glomerular filtration rate (eGFR) < 30 mL/min/1.73 m². Patients with gout symptoms or a history of uric acid treatment were also excluded. These exclusion criteria were applied to obtain a homogeneous cohort of uncomplicated essential hypertension so that any association between asymptomatic hyperuricemia and LV remodeling/function would be less confounded by comorbid cardiovascular or metabolic diseases. Then, according to whether they had HU (defined as more than 6 mg/dL [360 µmol/L] in females and more than 7 mg/dL [420 µmol/L] in males, according to Chinese expert consensus [15]), patients were divided into two groups: The essential hypertension with HU [hypertension (HU+)] group and the essential hypertension without HU [(hypertension HU-)] group. A retrospective control cohort (n = 62) was drawn from our CMR database, consisting of individuals referred for non‑specific symptoms (e.g., atypical chest discomfort, palpitations) or routine health evaluations who had normal blood pressure, normal laboratory profiles, and no structural cardiovascular or metabolic disease on assessment.

CMR protocol

CMR was performed using two 3.0T whole-body MRI scanners (Trio Tim and Magnetom Skyra, Siemens Medical Solutions, Erlangen, Germany). A manufacturer’s standard ECG-triggering device and the breath-hold technique were used during the entire examination, and data acquisition was performed during the breath-holding period. Localized imaging, consisting of imaging in the coronal, sagittal, and horizontal planes, was performed by using the True FISP sequence. A balanced steady-state free precession (bSSFP) sequence was used to acquire 8–12 continuous cine images from the mitral valve level to the LV apex in the short-axis view. Vertical LV 2- and 4-chamber long-axis view cine series were acquired as well. MRI scans were performed using the following parameters: repetition time [TR] of 2.81 ms or 3.4 ms, echo time [TE] of 1.22 ms, flip angle of 38° or 50°, slice thickness of 8 mm, field of view [FOV] of 360 × 300 mm² or 340 × 285 mm², and matrix of 208 × 139 or 256 × 166.

Image analysis

An experienced radiologist analyzed the CMR images of eligible subjects on an offline workstation (cvi42, v.5.10.2; Circle cardiovascular imaging, Calgary, Canada) without seeing the clinical data. The end-systolic and end-diastolic endocardium and epicardium on the short axis were drawn to obtain routine cardiac function indexes, including LV end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejection fraction (EF), and LV mass. Those indexes were standardized by body surface area (BSA). The LV remodeling index (LVRI) was calculated as LV mass/EDV.

The end-diastolic endocardium and epicardium of the short axis and two long-axis sections were drawn to analyze the LV strain parameters, including LV global circumferential peak strain (GCPS), global longitudinal peak strain (GLPS) and global radial peak strain (GRPS). These parameters represent the percentage change in length from the relaxed state to the contractile state when the myocardium moves in the radial, circumferential, and longitudinal directions. Since the myocardium shortening in the longitudinal and circumferential directions while thickening radially, the first two values are negative, while the latter is positive [16, 17].

The outlines of the LV endocardium and epicardium were first generated by the AI plug-in of cvi42, and then the radiologist made slight adjustments according to the most recent International Cardiac Magnetic Resonance Association guidelines [18].

Intra-observer and inter-observer reproducibility

The reproducibility of CMR LV strain parameters was assessed by two investigators. To determine the intra-observer variability, LV global strain parameters in 30 randomly selected cases (10 in each group) were reanalyzed after an interval of 1 month by the same radiologist. To determine the inter-observer variability, another investigator reanalyzed the images of the 30 subjects. Each observer was blinded to the subjects’ status and the findings of the other observer during the variability assessment.

Statistical analysis

Statistical analyses were performed with IBM SPSS (version 22.0, IBM SPSS Inc., Armonk, New York, US). All continuous variables were checked for normality using the Kolmogorov-Smirnov test. Variables are presented as the mean ± standard deviation. The disease duration with a large degree of skewness was presented by median (interquartile range). One-way analysis of variance (one-way ANOVA) and the Kruskal-Wallis test were used to compare the baseline characteristics among the control, hypertension (HU-) and hypertension (HU+) groups. One-way ANOVA was used when the data conformed to the homogeneity of variance and normal distribution assumptions, followed by an intergroup comparison using the Student–Newman–Keuls test. The Kruskal-Wallis test was used when the data exhibited skewed distributions. The relationships between SUA and LV myocardial strain and morphological parameters were analyzed by Pearson correlation. The input multivariable linear regression analyses were used to identify associations of CMR parameters and the basic clinical data. All candidate covariates were first assessed in univariable analyses. Variables demonstrating p < 0.10 were eligible for multivariable entry. Collinearity was evaluated using pairwise Pearson correlations and variance‑inflation factors (VIF); when r > 0.80 or VIF > 5, only the clinically more relevant variable was retained. Due to physiological differences between males and females, correlation analysis was conducted separately between different genders. Because the minus sign of GLPS and GCPS in LV myocardial strain represents the direction of myocardial motion from end-diastole to end-systole, their absolute values were used in the correlation analysis. Inter- and intraobserver agreements were assessed by evaluating intraclass correlation coefficients (ICCs) and Bland– Altman plots. A two-tailed P value < 0.05 was considered significant for all statistical tests.

Results

Study population and clinical baseline characteristics

Ultimately, 306 individuals, comprising 146 hypertension (HU−) patients, 98 hypertension (HU+) patients, and 62 healthy controls, were included in the study. The comparison of basic characteristics among the control, hypertension (HU-), and hypertension (HU+) groups are presented in Table 1. In addition to blood pressure, hypertensive patients had higher BMI and BSA than the control group. Moreover, there was no statistical difference in the course of hypertension between the hypertension groups (10.76 ± 7.46 vs. 10.87 ± 6.40, P = 0.909). SUA in the hypertension (HU+) group was significantly higher than the hypertension (HU-) group. The hypertension (HU+) group had a significantly lower mean eGFR compared to both the controls and the HU− hypertension group (P < 0.05).

Table 1.

Basic characteristics of controls and essential hypertension patients without HU and with HU

	Controls (n = 62)	Hypertension patients
		HU- (n = 146)	HU+ (n = 98)
Demographics
Age (years)	53.91 ± 10.37	55.59 ± 12.64	54.08 ± 13.23
Female, n (%)	27(43.5)	68 (46.6)	39 (40)
BMI (kg/m²)	23.08 ± 2.87	24.25 ± 2.84*	25.69 ± 3.11*#
BSA (m²)	1.65 ± 0.16	1.68 ± 0.16	1.76 ± 0.18*#
Hemodynamic variables
Heart rate (beats/min)	71.61 ± 9.71	72.34 ± 13.75	74.73 ± 13.05
SBP (mmHg)	113 ± 10.17	139.77 ± 18.51*	139.09 ± 22.19*
DBP (mmHg)	71.81 ± 7.84	84.98 ± 13.63*	87.11 ± 17.48*
Laboratory data
SUA (mg/dL)	295.73 ± 69.9	319.86 ± 56.96	467.73 ± 74.05*#
Creatinine (mg/dL)	76.14 ± 19.62	73.32 ± 21.02	81.77 ± 23.98#
Cys-C (mg/dL)	0.89 ± 0.18	0.95 ± 0.22	1 ± 0.22*
TG (mmol/L)	1.58 ± 1.27	1.49 ± 0.99	1.81 ± 0.94
Total cholesterol (mmol/L)	4.26 ± 0.8	4.36 ± 1.06	4.27 ± 1.01
HDL (mmol/L)	1.38 ± 0.4	1.31 ± 0.41	1.22 ± 0.35*
LDL (mmol/L)	2.42 ± 0.76	2.52 ± 0.92	2.51 ± 0.85
eGFR (mL/min/1.73 m2)	95.19 ± 18.02	91.69 ± 19.9	84.66 ± 22.99*
blood glucose (mmol/L)	5.22 ± 0.83	5.64 ± 1.07	5.65 ± 1.03
Hypertension treatment
HTN duration (years)	-	6.8 ± 7.9	5.7 ± 5.6
ACEI/ARB, n (%)	-	28 (40)	15 (35)
Beta-blocker, n (%)	-	28 (40)	16 (40)
Calcium channel blocker, n (%)	-	40 (55.7)	20 (47.5)
Diuretics, n (%)	-	11 (15)	6 (15)
Hypertension duration (years)	-	10.76 ± 7.46	10.87 ± 6.40
antihypertensive therapy, n (%)	-	120 (82.2)	84 (85.7)
ACEI/ARB, n (%)	-	67 (45.9)	46 (46.9)
Beta-blocker, n (%)	-	47 (32.2)	34 (34.6)
Calcium channel blocker, n (%)	-	53 (36.3)	33 (33.6)
Diuretics, n (%)	-	17 (11.6)	13 (13.2)

The values are the mean ± SD. Numbers in the brackets are percentages

BMI Body mass index, BSA Body surface area, Cys-C Cystatin C, DBP Diastolic pressure, eGFR Estimated glomerular filtration rate, HDL High-density lipoprotein cholesterol, HU Hyperuricemia, LDL Low-density lipoprotein cholesterol, SBP Systolic pressure, SUA Serum uric acid, TG Plasma triglycerides, ACEI Angiotensin-converting enzyme inhibitor, ARB Angiotensin receptor blocker. Percentages for medication classes refer to the proportion of patients receiving each class; totals exceed 100% because many patients used ≥ 2 antihypertensive agents

* P < 0.05 versus controls

# P < 0.05 versus HU - group

Comparison of CMR parameters among the three groups

The LV function and structure findings for the observed groups are shown in Table 2. There was no significant difference in routine CMR parameters, such as EF and SV, among the three groups. LV GLPS deteriorated from the control group to the hypertension (HU−) group to the hypertension (HU+) group (P < 0.05). LV GCPS and GRPS were lower in the hypertension (HU+) group than in the control group and hypertension (HU-) group. Fig. 1 shows representative CMR cine images and CMR-derived peak strain curves in a normal control, a hypertensive patient without HU, and a hypertensive patient with HU, whose age, sex and BMI were matched. The LVMI and LVRI in both the hypertension (HU−) and hypertension (HU+) groups were significantly higher than that of the control group (all P < 0.001). In patients with hypertension, the LVMI and LVRI of the hypertension (HU+) group were larger than that of the hypertension (HU-) group. Fig. 2 demonstrated that LVMI rose further with hyperuricemia in male hypertensive patients, whereas in female patients LVMI was similar between HU − and HU+ groups.

Table 2.

Comparisons of CMR findings among controls, hypertension (HU −) group and hypertension (HU +) group

	Controls (n = 62)	Hypertension patients		P value
		HU- (n = 146)	HU་ (n = 98)
CMR-derived cardiac geometric and functional parameters
LVEF (%)	63.64 ± 7.66	63.79 ± 10.14	61.48 ± 12.57	0.213
LVEDVI (mL/m²)	78.56 ± 15.23	81.29 ± 21.23	82.19 ± 22.24	0.539
LVESVI (mL/m²)	28.79 ± 8.78	31.09 ± 17.84	33.32 ± 18.85	0.242
LVSVI (mL/m²)	49.77 ± 10.22	50.48 ± 11.37	49.22 ± 12.59	0.702
LVMI (g/m2)	52.66 ± 18.08	62.05 ± 18.23*	68.95 ± 23.78*#	< 0.001
LVRI (g/ml)	0.69 ± 0.28	0.77 ± 0.16*	0.84 ± 0.24*#	< 0.001
CMR strain of LV (%)
GRPS	36.64 ± 8.53	34.16 ± 10.44	30.87 ± 13*	0.004
GCPS	-21.04 ± 2.59	-20.27 ± 4.21	-18.71 ± 5.55*#	0.002
GLPS	-14.31 ± 2.05	-12.85 ± 3.11*	-10.63 ± 4.01*#	< 0.001

LV Left ventricular, EF Ejection fraction, EDV End diastolic volume, ESV End systolic volume, HU Hyperuricemia, SV Stroke volume, M Mass, I Indexed to BSA, LVRI Left ventricular remodeling index, GRPS Global radial peak strain, GCPS Global circumferential peak strain, GLPS Global longitudinal peak strain

* P < 0.05 versus controls

#P < 0.05 versus HU (-) group

Fig. 1.

Representative CMR pseudocolor images at end‑systole and CMR‑derived peak‑strain curves in a normotensive control (A), a hypertensive patient without hyperuricemia (HU−, B) and a hypertensive patient with hyperuricemia (HU+, C), matched for age, sex and BSA. A1–C1 Horizontal 4‑chamber long‑axis strain maps; a1–c1 global longitudinal peak‑strain curves. A2–C2 Short‑axis strain maps; a2–c2 global circumferential peak‑strain curves. Colour scale (right): blue shades indicate larger absolute myocardial shortening (up to ≈ − 20%), whereas blue–green–yellow denote progressively lower strain. GCPS, global circumferential peak strain; GLPS, global longitudinal peak strain

Fig. 2.

Sex‑stratified box‑and‑whisker plots of left ventricular peak strains and mass index in controls and hypertensive patients with (HU+) and without (HU−) hyperuricemia. GRPS, global radial peak strain; GCPS, global circumferential peak strain; GLPS, global longitudinal peak strain; LVMI, left ventricular mass index; LVRI, left ventricular remodeling index. * P < 0.05

Correlations between SUA and LV strain and morphological parameters

As shown in Fig. 3, in patients with essential hypertension, there was a significant linear correlation between SUA and the absolute values of LV global myocardial strain in both females and males (Females: GCPS, r = -0.22, P = 0.02; GLPS, r = -0.28, P = 0.003. Male: GRPS, r = -0.24, P = 0.005; GCPS, r = -0.25, P = 0.003; GLPS, r = -0.39, P < 0.001). In terms of morphology, LVMI (r = 0.34, P < 0.001) and LVRI (r = 0.24, P < 0.001) were linearly correlated with SUA in males, but not in females.

Fig. 3.

The relationship between LV global peak strain and morphological parameters and SUA in patients with essential hypertension of different genders. SUA, serum uric acid; LV, left ventricular; GRPS, global radial peak strain; GCPS, global circumferential peak strain; GLPS, global longitudinal peak strain; LVMI, left ventricular mass index; LVRI, left ventricular remodeling index

Associations of CMR imaging variables with clinical factors

All candidate variables for linear regression analysis were selected according to the clinical basis and the results of univariate analysis (Supplementary Table S1). As shown in Table 3, after adjusting for age, BSA, and clinical confounding factors, SUA was associated with LV GLPS regardless of gender (β = -0.263, P = 0.025 in females and β = -0.328, P = 0.001 in males). In terms of myocardial structure,, as shown in Table 4, after considering age and clinical confounding factors, it was found that SUA was associated with LVMI (β = 0.189, P = 0.03) in males, but not in females. The association of SUA with GLPS (both sexes) and with LVMI in males remained significant after additional adjustment for medication class (Supplementary Table S2-3).

Table 3.

Multivariable linear analysis of LV strains index in hypertensive patients of different genders

	Females						Males
	GRPS (R² = 0.042)		Absolute value of GCPS (R² = 0.080)		Absolute value of GLPS (R² = 0.121)		GRPS (R² = 0.296)		Absolute value of GCPS (R² = 0.185)		Absolute value of GLPS (R² = 0.253)
	β	P	β	P	β	P	β	P	β	P	β	P
SUA	-0.197	0.061	-0.183	0.093	-0.263*	0.025	-0.133	0.189	-0.172	0.082	-0.328*	0.001
Age	-0.027	0.790	0.023	0.830	0.044	0.716	-0.114	0.405	-0.160	0.228	0.023	0.864
BSA	0.091	0.379	0.146	0.171	0.003	0.978	-0.181	0079	-0.141	0.159	-0.159	0.094
DBP	-	-	-	-	-	-	-	-	-	-	0.018	0.855
Creatinine	-	-	-0.171	0.134	-	-	-0.046	0.752	-0.044	0.795	-0.096	0.489
TG	-	-	-	-	-0.068	0.530	-	-	-	-	-	-
Total cholesterol	-	-	-	-	-	-	0.268	0.953	0.421	0.127	-	-
Cys-C	-	-	-	-	-0.080	0.525	-0.139	0.259	-0.174	0.147	-0.178	0.117
HDL	-	-	-	-	0.100	0.361	-	-	-	-	0.016	0.863
LDL	-	-	-	-	-	-	-0.149	0.601	-0.270	0.330	-	-
eGFR#	-	-	-	-	-	-	0.032	0.838	0.029	0.953	-0.024	0.874

Abbreviation of BP, BSA, Cys-C, HDL, LDL, SUA and TG are shown in Table 1, and LV, GRPS, GCPS, GLPS, LVMI, LVRI in Table 2. β, Standard regression coefficient. Variables for multivariable model were selected on clinical grounds, guided by univariable correlation with P value < 0.10 and the absence of collinearity

# eGFR did not enter the female strain model because it was non‑significant in univariable testing. A sensitivity model including eGFR instead of cystatin C is shown in Supplementary Table S2 and yielded similar results

* P < 0.05

Table 4.

Multivariable linear analysis of LVMI and LVRI in hypertensive patients of different genders

	Females				Males
	LVMI (R² = 0.113)		LVRI (R² = 0.159)		LVMI (R² = 0.343)		LVRI (R² = 0.183)
	β	P	β	P	β	P	β	P
SUA	-0.042	0.696	0.044	0.701	0.189*	0.031	0.099	0.299
Age	-0.058	0.576	-0.074	0.547	-0.113	0.301	-0.084	0.481
BMI	0.014	0.897	0.103	0.356	0.104	0.236	0.044	0.648
SBP	0.262*	0.011	-	-	0.190*	0.021	0.191*	0.036
Creatinine	-	-	0.019	0.872	0.054	0.652	0.088	0.468
Total cholesterol	-	-	-0.226*	0.039	-0.173*	0.038	-	-
Cys-C	-	-	0.230	0.089	0.233*	0.031	-	-
eGFR	-	-	-	-	-0.214	0.099	-0.415*	0.003
blood glucose	-0.150	0.144	-	-	-	-	-	-

Abbreviation of BMI, HDL, Cys-C, SBP, SUA and eGFR are shown in Table 1, and LVMI, LVRI in Table 2. β, Standard regression coefficient. Variables for multivariable model were selected on clinical grounds, guided by univariable correlation with P value < 0.10 and the absence of collinearity

* P < 0.05

Reproducibility of CMR feature tracking

As shown in Table 5; Fig. 4, the intra- and interobserver agreements in the measurement of global LV peak strain were excellent (ICC = 0.925 ~ 0.958 and 0.891 ~ 0.929, respectively).

Table 5.

Inter- and intraobserver variability of tissue tracking

	Intraobserver		Interobserver
	ICC	95% CI	ICC	95% CI
GLPS	0.925	0.878–0.963	0.891	0.815–0.959
GCPS	0.958	0.905–0.984	0.912	0.886–0.942
GRPS	0.937	0.883–0.975	0.929	0.852–0.958

ICC Intraclass correlation coefficient, CI Confidence interval; definitions of GLPS, GRPS, GCPS are shown in Table 2

All P < 0.01

Fig. 4.

Bland–Altman plots with limits of agreement (95% confidence intervals) demonstrating the excellent intra-observer and inter-observer reproducibility of CMR-FT parameters. Solid lines represent bias and dotted lines represent 95% limits of agreement. GRPS, global radial peak strain; GCPS, global circumferential peak strain; GLPS, global longitudinal peak strain

Discussion

The adverse cardiovascular events caused by HU have been a concern for a long time [8, 19, 20]. Epidemiological and animal studies have suggested that SUA levels are associated with blood pressure and may predict the onset and progression of hypertension [21, 22]. However, whether SUA levels contribute to subsequent target organ damage in hypertensive patients remains controversial [23, 24]. From the perspective of CMR, we retrospectively analyzed the possible effect of SUA on LV function and structure in patients with essential hypertension. The main findings are as follows: First, HU and hypertension coexisting are related to further changes in LV function and morphology. Second, SUA levels were closely related to myocardial strain, particularly longitudinal strain. Third, the impact of hyperuricemia on LV structure may differ between genders.

In hypertensive patients without HU, our data show that the LV GLPS of hypertensive patients was lower than that of the control group, which is consistent with the data of previous echocardiographic studies [25].With the progress of scanning and postprocessing technology, CMR has gradually become the gold standard for evaluating cardiac strain [26]. Although HU often coexists with hypertension, there seems to be no previous CMR study on the effects of HU on cardiac function and deformation in hypertensive patients. Through this study, we found that the absolute values of LV radial, circumferential and longitudinal peak strain in the hypertension HU+ group were significantly lower than those in normal subjects and that GCPS and GLPS were worse than those in the hypertension HU- group. In the past, some epidemiological and systematic reviews showed that HU may moderately increase the traditional risk factors such as blood pressure, so the 2018 European Guidelines for the Management of Hypertension included SUA in routine workups [7, 27–29]. However, because the increase in uric acid is often accompanied by other metabolic disorders, the effect of uric acid on the heart was often confusing. The American Rheumatic Society (ACR) issued gout management guidelines to recommend that patients with asymptomatic HU should not use any uric acid-lowering drugs if possible but did not elaborate on what to do for SUA increases related to other metabolic diseases [30]. In our study, the hypertensive patients with HU exhibited a further decrease in cardiac strain compared to those without HU, despite similar levels of blood pressure and BMI. This suggests that attention should be paid to the SUA in hypertensive patients. Despite excluding advanced renal disease, HU+ patients showed significantly lower eGFR than controls and HU− group, underscoring the close linkage between hyperuricemia and early renal functional decline. Consistent with the concept that systolic strain impairment often precedes overt hypertrophy [31], renal function markers predicted LVMI but not GLPS. Sensitivity analyses that forced eGFR into the female strain model did not alter this pattern (Supplementary Table S4).

According to the results of the study, we found a certain correlation between SUA and LV myocardial strain, suggesting that LV function may progressively deteriorate with increasing SUA levels. Although the SUA–strain correlations are modest (|r| ≈ 0.22–0.39,), such effect sizes are typical for single metabolic factors in multifactorial cardiovascular traits. After adjusting for confounding factors, the correlation between SUA and LVGLPS remained statistically significant. We speculate that this is related to the geometric factors of myocardial strain. The lower R² values seen in models with GRPS are not unexpected, as GRPS captures heterogeneous wall‑thickening and through‑plane motion and is therefore intrinsically more variable. By contrast, past studies have confirmed that GLPS and GCPS are related to the development and prognosis of many cardiovascular diseases and that GLPS damage usually occurs first [32, 33]. In addition, GLPS and GCPS mainly reflect the contractile function of longitudinal and circumferential fibers of the LV, respectively. Compared with annular fibers, subendocardial longitudinal fibers are more vulnerable to ischemia and hypoxia because of their higher stress and higher oxygen demand [12, 34]. Therefore, GLPS may be more sensitive to changes in cardiac function associated with elevated uric acid. Our findings extend the scant CMR data by showing that, in essential hypertension, elevated SUA is also related to LV strain impairment, and are concordant with speckle‑tracking results reported in diabetic cohorts [35].

In addition to the decrease in myocardial strain, our results also show further changes in LV structure in patients with HU. Compared with those in the hypertension HU- group, the LV mass and LVRI increased in the hypertension HU+ group. At present, the pathophysiological mechanisms of LV remodeling in patients with hypertension can be divided into two types. In addition to the direct mechanical mechanism of compensatory myocardial hypertrophy caused by increased load, biological humoral mechanisms such as the renin-angiotensin-aldosterone system, sympathetic autonomic nervous system and increased oxidative stress also play an important role in ventricular remodeling [36, 37]. These metabolic mechanisms are related to myocardial interstitial fibrosis, which mainly leads to decreased LV elasticity, LV deformation, diastolic dysfunction and hypertrophy [2, 38]. Some experiments have shown that SUA can inhibit the production of nitric oxide, stimulate the release of inflammatory mediators, and directly lead to endothelial dysfunction and vascular smooth muscle cell proliferation [39–41]. The lack of significant difference in blood pressure between hypertensive patients with and without HU suggests that other factors may be involved in the further increase of LVMI observed in patients with HU. Our results indicate a potential role of SUA in the humoral regulation mechanism of ventricular remodeling. The relatively modest R² values, particularly in the univariable models, signify that a large proportion of variance in myocardial strain and mass remains unexplained, likely reflecting factors we could not measure—such as dietary sodium or fructose intake, long‑term blood‑pressure variability and genetic predisposition.

In the past, there has been some controversy about the effect of uric acid on the heart in different genders [42]. Our multivariable analysis demonstrated an independent association between SUA and myocardial strain in both males and females. However, SUA was significantly linked to left ventricular (LV) remodeling only in males. The relationships persisted when antihypertensive drug classes were included in sensitivity models, indicating that treatment regimen did not materially confound the findings. These findings align with previous studies conducted in Japan, which reported a similar gender-specific association [43, 44]. In contrast, several large-scale epidemiological investigations in Europe have identified a stronger correlation between SUA levels and cardiovascular events in women than in men [45–48]. Interestingly, these conclusions differ from recent findings in a Chinese cohort, where males appeared to be more affected by elevated SUA [49]. These apparent discrepancies may be partially explained by differences in study populations, including genetic, environmental, and lifestyle factors. Additionally, the interplay between SUA and other metabolic factors, such as oxidative stress, inflammation, and hormonal regulation, is likely to influence myocardial remodeling. While myocardial hypertrophy is widely recognized as a predictor of cardiovascular events, it also represents a compensatory response to increased cardiac workload and metabolic stress [37]. This dual nature of myocardial remodeling underscores the complexity of SUA’s role in cardiovascular pathophysiology.

The gender-specific differences observed in our study might also be attributed to the varying hormonal and metabolic environments between males and females. Several mechanisms may account for this divergence. First, estrogen promotes uricosuria and augments nitric‑oxide bioavailability, potentially attenuating uric‑acid–driven oxidative stress and RAAS activation in women [50]. Second, experimental and clinical studies indicate that males experience a higher oxidative‑stress burden, which can amplify hypertrophic responses to metabolic stimuli such as SUA [51]. Finally, meta‑analytic data suggest the SUA–LVH relationship itself is modulated by sex, supporting a biological interaction [42]. Future research should focus on prospective, multicenter studies that account for genetic and hormonal influences on SUA-related cardiac remodeling. Furthermore, mechanistic studies exploring the molecular pathways linking SUA to myocardial strain and hypertrophy could provide valuable insights.

This study has several limitations. First, its retrospective, cross‑sectional, single‑centre design may introduce selection bias and cannot establish causality. Prospective trials are warranted to determine whether pharmacological urate‑lowering regimens can attenuate cardiac remodeling and improve clinical outcomes. Second, although we excluded overt heart failure, coronary disease, diabetes and advanced CKD to obtain a homogeneous hypertensive cohort, doing so—together with the absence of uniform data on lifestyle factors (diet, alcohol, smoking, physical activity) or genetic background—limits generalisability and leaves residual confounding possible. Third, the sample size is moderate and drawn from a single, ethnically homogeneous population, which may restrict external validity. Finally, we evaluated global strain and LV mass only; diastolic indices and tissue characterisation (e.g., T1 mapping, LGE) were not available and should be explored prospectively.

Conclusions

The coexistence of HU may suggest further reductions in LV function and alterations in LV structure in patients with essential hypertension. Elevated serum uric acid levels correlate with myocardial strain, particularly longitudinal strain. In essential hypertension, asymptomatic hyperuricemia correlates with subclinical systolic dysfunction in both men and women, but contributes to concentric LV hypertrophy only in men, underscoring the need for sex‑specific risk assessment. Routine SUA assessment may aid risk stratification—particularly in hypertensive men—in addition to traditional cardiovascular risk factors.

Abbreviations

BMI

Body mass index

BSA

Body surface area

CMR

Cardiac magnetic resonance

EDV

End-diastolic volume

EF

Ejection fraction

eGFR

Estimated glomerular filtration rate

ESV

End-systolic volume

GCPS

Global circumferential peak strain

GLPS

Global longitudinal peak strain

GRPS

Global radial peak strain

HDL

High-density lipoprotein cholesterol

HU

Hyperuricemia

LDL

Low-density lipoprotein cholesterol

LV

Left ventricular

SUA

Serum uric acid

SV

Stroke volume

TG

Triglycerides

Authors’ contributions

WFY, ZGY and JW designed the study. WFY and XML analyzed the data and wrote the manuscript. SQY, YL and RX participated in the study design and review of the manuscript. ZGY and JW supervised the overall study and contributed to study design, editing and review of the manuscript. LJ, YG, JKL were responsible for collecting, sorting and statistical data. ZGY and JW are the guarantors of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors read and approved the final manuscript.

Funding

This work was supported by a grant from the 1·3·5 project for disciplines of excellence, West China Hospital, Sichuan University (ZYGD23019) and the Science and Technology Support Program of Sichuan Province (2022NSFSC0828, 2024NSFSC1795).

Data availability

The anonymised datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki. Ethics approval was obtained from the Ethics Committee on Biomedical Research, West China Hospital of Sichuan University (Approval No. 2019 − 811). Written informed consent was waived owing to the retrospective design and prior anonymisation of data, in line with the Chinese Measures for Ethical Review of Biomedical Research Involving Humans.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.