Left atrial reservoir strain and the left atrioventricular coupling index in association with left ventricular diastolic dysfunction in non-dialysis chronic kidney disease with preserved ejection fraction: a cross-sectional study

Left ventricular diastolic dysfunction (LVDD) is common in chronic kidney disease (CKD) despite preserved left ventricular ejection fraction (LVEF), whereas conventional echocardiographic grading may be challenging in this setting. We examined the patterns of left atrial reservoir strain (LASr) and the left atrioventricular coupling index (LACI) across CKD stages and their associations with echocardiography-defined LVDD in non-dialysis CKD patients with preserved LVEF. In this single-center cross-sectional study, non-dialysis CKD patients with LVEF ≥ 50% and sinus rhythm were consecutively enrolled between November 2024 and November 2025, together with healthy controls with a comparable age and sex distribution. CKD was classified by estimated glomerular filtration rate (eGFR) as G1–2 (≥ 60), G3 (30–59), and G4–5 (< 30 mL/min/1.73 m²). Phasic left atrial strain was measured by two-dimensional speckle-tracking echocardiography. LACI was calculated as minimal left atrial volume divided by left ventricular end-diastolic volume × 100%. LVDD was classified using a guideline-based approach, with a prespecified alternative strategy when tricuspid regurgitation velocity was not measurable. Multivariable logistic regression, receiver operating characteristic analysis, reclassification analysis, and decision-curve analysis were used to assess independent associations and incremental value. A total of 177 CKD patients and 50 controls were included. LASr progressively declined with worsening CKD (controls: 42.93 ± 5.62% vs. G4–5: 29.28 ± 6.10%; P for trend < 0.001), whereas LACI increased across CKD stages (P for trend = 0.003). LVDD was present in 39.0% of the CKD cohort and was more frequent in G4–5 disease (60.9%). After adjustment for age, systolic blood pressure, hemoglobin, eGFR, and left ventricular mass index, LASr was independently associated with lower odds of LVDD (per 1 SD increase: OR 0.455, 95% CI 0.290–0.714), whereas LACI was independently associated with higher odds of LVDD (per 1 SD increase: OR 3.156, 95% CI 1.914–5.204). Adding LASr and then LACI to the clinical base model improved discrimination for LVDD (AUC: 0.763, 0.821, and 0.872, respectively; both DeLong P < 0.05 vs. the preceding model). Reclassification analyses and decision-curve analysis also supported incremental value. The main findings remained directionally consistent after additional adjustment for diuretic use and after LVDD was reclassified according to the 2025 ASE update. In non-dialysis CKD patients with preserved LVEF, lower LASr and higher LACI were associated with echocardiography-defined LVDD. LASr and LACI may provide complementary information on left atrial dysfunction and atrioventricular volumetric alteration and may improve identification of LVDD when added to a clinical model.