Renal microvascular pulsatility by superb microvascular imaging predicts post-discharge outcomes in acute heart failure
K Kayama, T Sugimoto, R Kudo, Y Sakurai, Y Sekiya, T Okada, T Sawazaki, Y SeoAbstract
Background
Renal dysfunction in acute heart failure is associated with poor prognosis. Although renal congestion has gained attention, it remains difficult to distinguish congestion from low perfusion using routine bedside assessment. Superb microvascular imaging (SMI) uses clutter suppression to attenuate motion artefacts, enabling contrast-free visualisation of low-velocity renal microvascular flow and quantification using the vascular index (VI).
Purpose
To investigate the prognostic value of an SMI-derived index of cyclic microvascular variation at discharge for clinical outcomes.
Methods
We prospectively enrolled 140 consecutive patients hospitalised as an emergency for acute heart failure between 2023 and 2025 who were discharged alive (age 80±11 years; 60% male). At discharge, blood and urine biochemistry, transthoracic echocardiography, renal venous Doppler, and renal ultrasound with SMI were performed. VI was defined as the percentage of flow-signal area within predefined regions of interest in both the renal cortex and medulla (Figure1A). Pulsus VI was defined as the difference between maximum and minimum VI within a cardiac cycle, averaged over three cardiac cycles (Figure1B). The primary endpoint was a composite of all-cause death or heart-failure rehospitalisation. Discrimination was assessed by ROC analysis, and time-to-event analyses by Kaplan–Meier and Cox models.
Results
Over 377±272 days of follow-up, 23 patients experienced the primary endpoint. Patients with events had lower use of sodium–glucose cotransporter-2 inhibitors and lower haemoglobin and eGFR, and higher urinary beta2-microglobulin, liver-type fatty acid-binding protein, and albumin-to-creatinine ratio. No significant differences were observed in echocardiographic parameters or renal venous Doppler patterns. Pulsus VI was significantly higher in patients with events. Pulsus VI predicted the primary endpoint (AUC 0.662). Patients with pulsus VI above a ROC-derived cut-off (>0.145) had higher event risk (log-rank p<0.01; unadjusted HR 3.4, 95% CI 1.40-8.30) (Figure2). In multivariable Cox models, pulsus VI above cut-off remained associated with the endpoint after adjustment for haemoglobin, eGFR and BNP (model 1: adjusted HR 2.8, 95% CI 1.08-7.09, p=0.03) and after adjustment for urinary liver-type fatty acid-binding protein, albumin-to-creatinine ratio and beta2-microglobulin (model 2: adjusted HR 3.4, 95% CI 1.27-9.26, p=0.01).
Conclusion
Renal microvascular pulsatility quantified by SMI at discharge was associated with post-discharge outcomes in acute heart failure, supporting its potential role as a non-invasive tool for risk stratification beyond conventional renal function measures.