The goal of this study is to quantitatively assess renal microcirculation changes by contrast-enhanced ultrasound and to obtain systemic hemodynamic information by ultrasound Doppler at the same time, to analyze the relationship between renal microcirculation changes and systemic hemodynamic changes, and to explore its predictive value in renal function recovery in patients with critical acute kidney injury. The main questions it aims to answer are:

1. To explore the quantitative parameters of contrast-enhanced ultrasound which can reflect the changes of renal microcirculation.
2. To explore the relationship between renal microcirculation and systemic hemodynamics.
3. To explore the value of renal microflow changes quantitatively evaluated by contrast-enhanced ultrasound in predicting renal function recovery.

Quantitative analysis of cortical perfusion by contrast-enhanced ultrasound； Echocardiography； Quantitative results of renal artery and vein spectrum

Inclusion Criteria:

• Age >18 years old;
• AKI was diagnosed < 24 hours after first admission to intensive care unit (ICU) and the expected length of stay in ICU ≥48 hours;
• Meet Kidney Disease Improving Global Outcomes (KDIGO) 2012 diagnostic criteria for acute kidney injury;
• Contrast-enhanced ultrasound was performed within 24 hours after diagnosis of acute kidney injury (AKI).

Exclusion Criteria:

• Known severe chronic kidney disease (CKD≥ stage 4) or undergoing hemodialysis treatment;
• Kidney transplantation or renal malignancy;
• Terminal stage of malignant tumor;
• Pregnancy;
• Renal artery stenosis or renal vein thrombosis;
• Allergy to contrast agent SonoVue(BraccoSpA, Milan, Italy) or its components, or the presence of severe cardiopulmonary insufficiency (e.g., right-to-left shunt heart disease or pulmonary systolic blood pressure >90mmHg).

Published contrast-enhanced ultrasound data from critically ill patients are limited. Through literature review, the area under the curve of contrast-enhanced ultrasound to predict kidney recovery in critically ill AKI patients is 0.72, and the width of confidence interval is 0.35. When α=0.05, the sample size N=34 is obtained by using PASS software. Considering the 20% shedding rate, 44 patients are expected to be enrolled in this study.