Objectives: Diagnosing persistent acute kidney injury (AKI) as opposed to transient AKI in critically ill patients may help physicians in making treatment decisions. This diagnosis relies chiefly on urinary indices, which may be of limited value or difficult to obtain. We assessed the performance of the Doppler renal resistive index (RI) in diagnosing persistent AKI.
Design: Prospective observational study.
Setting: Twenty-four-bed medical intensive care unit in a university hospital.
Patients: Consecutive patients requiring mechanical ventilation, without severe chronic renal dysfunction or receiving diuretic therapy.
Definitions: Persistent AKI was defined as AKI lasting longer than 3 days. AKI resolving within 3 days in a patient with a cause of renal hypoperfusion was considered to be transient AKI. Results are reported as median values with interquartile range (IQR).
Measurements and main results: Of the 51 patients enrolled in the study, 16 had no AKI, 13 had transient AKI, and 22 had persistent AKI. The RI was 0.71 (0.66-0.77) in the no-AKI group, 0.71 (0.62-0.77) in the transient AKI group, and 0.82 (0.80-0.89) in the persistent AKI group (P < 0.0001). The RI was better than urinary indices for diagnosing persistent AKI. The area under the RI ROC curve was 0.91 [95% confidence interval (95% CI) 0.83-0.99; P < 0.0001]. An RI > 0.795 had a 92% sensitivity and 85% specificity for persistent AKI. Logistic regression analysis revealed that an RI > 0.795 [odds ratio (OR) 28.2; 95% CI 4.0-198] and a higher logistic organ dysfunction score (OR 1.85/point; 95% CI 1.20-2.85) predicted persistent AKI.
Conclusion: These preliminary results suggest that Doppler renal RI may be a promising tool for predicting the reversibility of AKI in critically ill patients.