Introduction
Omega-3 (ω-3) and omega-6 (ω-6) unsaturated fatty acids are associated with cardiovascular disease¹ and worse cardiovascular outcomes². Previous studies showed no statistically significant associations of plasma ω-6 or ω-3 levels with left ventricular (LV) structural parameters by using cardiac magnetic resonance imaging (CMR)³. However, no studies have explored the effects of uric acid (UA) and abnormal ω-6/ω-3 values on left heart strain parameters. Therefore, this study investigated the effects of UA and abnormal ω-6/ω-3 values on left heart function by cardiac magnetic resonance feature tracking (CMR-FT).

Methods
This study included 300 patients with normal left ventricular ejection fraction (LVEF) who underwent CMR from September 2019 to May 2022 in our cardiology department. All subjects gave written informed consent. They were divided into two groups (ω6/ω3>5 and ω6/ω3≤5), and each of the groups was divided into four independent subgroups using UA quartiles. The exclusion criteria included LVEF<50%, acute renal insufficiency, malignant tumor, and presence of organic heart disease. All individuals underwent CMR on a 3T MR scanner (MAGNETOM Skyra, Siemens Healthineers AG, Erlangen, Germany). Imaging parameters for TrueFISP cine were as follows: TR=39.2 ms, TE=1.43 ms, flip angle=39°, slice thickness=8 mm, matrix size=208*139, voxel size=1.6*1.6*6 mm³, acquisition time=8s, acquisition heartbeat=4.62, and FOV=234 mm*280 mm. It was performed from the base to the apex level on short-axis and long-axis views; continuous cine imaging results were subjected to post-processing. CMR images of patients were utilized to evaluate structure, function, and myocardial strain parameters in the left atrium (LA) and left ventricle (LV) via CVI42 software (Circle Cardiovascular Imaging, Inc., Calgary, Canada) (Figure 1). All data were analyzed using SPSS statistical software (version 26.0; SPSS Inc., Chicago, IL, USA). Normally distributed continuous variables were expressed as means ± standard deviations and compared using Student’s t-test or one-way analysis of variance; non-normally distributed variables were expressed as medians (interquartile ranges) and compared using the Mann–Whitney U test or Kruskal–Wallis test. To identify independent indicators of LV strain, variables with P<0.05 in univariate linear regression analyses were included in multivariate linear regression models using stepwise forward selection.

Results
Compared with the ω6/ω3≤5 group, LV global longitudinal strain (LV-GLS), LS-apical, LV global radial strain (LV-GRS), RS-apical, circumferential strain-apical (CS-apical), and CS-mid were increased in the ω6/ω3>5 group (P<0.05) (Table 1). At the same UA quartile, ω6/ω3 values had no effect on LA or LV strain (P>0.05). In the ω6/ω3>5 group, LV-GLS, LS-apical, LV-GRS, RS-apical, RS-mid, RS-basal, LV global circumferential strain (LV-GCS), CS-apical, CS-mid, CS-basal, and peak diastolic strain rate (PDSR-L/S)-longitudinal/circumferential progressively decreased with increasing UA quartile (P<0.05). However, there were no significant differences in LA reservoir function, conduit function, and pump function among subgroups with different UA quartiles (P>0.05) (Table 2). Univariate linear regression demonstrated that hyperuricemia was associated with impaired LV strain (GLS, GRS, GCS) (P<0.05) (Table 3). Adjusted multivariate linear regression analyses showed that hyperuricemia had an independent effect on impaired LV strain (GRS, GCS) (P<0.05). Furthermore, triglycerides (TG), left atrial ejection fraction (LAEF), female sex, and presence of hypertension were independently associated with LV-GLS; glycated hemoglobin (HbA1C), female sex, LAEF, and LA early negative peak strain rate (LA-SRe) were independently associated with LV-GRS; and LAEF, HbA1C, low-density lipoprotein cholesterol (LDL-C), female sex, and LA-SRe were independently associated with LV-GCS (Table 4).

Discussion and Conclusion
In this study, we used magnetic resonance feature tracking to examine the effect of UA level on left heart function at multiple ω6/ω3 values. We found that an increase in UA could lead to LV strain reduction in patients with ω6/ω3>5, predominantly in the apical region, although it did not have a significant effect on LA strain. Furthermore, multivariable linear regression analysis indicated that hyperuricemia was an independent indicator of LV strain. These findings suggest a unique role for LV strain in the assessment of altered left heart function, implying that high ω6/ω3 values may exacerbate LV dysfunction in patients with hyperuricemia. Our results will help to improve the clinical management and treatment of patients with hyperuricemia.

Acknowledgements

We thank the Siemens Research Team and the scientific adviser, Xiaofeng Qu, for organizational guidance during this project. Their contributions considerably improved the usefulness of the results.