From the Guidelines
Acute Kidney Injury (AKI) should be diagnosed when the serum creatinine increases by ≥0.3 mg/dL within 48 hours or is ≥50% from baseline or when the urine output is reduced below 0.5 mL/kg/h for >6 hours.
Diagnostic Algorithm
The diagnostic algorithm for AKI in inpatients involves monitoring serum creatinine levels and urine output.
- Serum creatinine: an increase of ≥0.3 mg/dL within 48 hours or ≥50% from baseline indicates AKI 1.
- Urine output: a reduction below 0.5 mL/kg/h for >6 hours indicates AKI 1.
Limitations of Traditional Biomarkers
Traditional biomarkers such as serum creatinine and urine output have limitations, including delayed changes following kidney injury and low sensitivity and specificity 1.
Electronic Alerts
Electronic alerts can be used to detect AKI, with algorithms operating in laboratory information management systems or hospital information systems 1.
- A delta check looking for any absolute rise of 0.3 mg/dL within a certain time period can detect large numbers of AKI patients.
- More sophisticated alert algorithms can interface with the Patient Administration System to identify dialysis patients and prevent false-positive alerts.
Guideline Recommendations
The KDIGO guidelines provide recommendations for the diagnosis and management of AKI, but may require interpretation for individual patients with specific clinical circumstances 1.
From the Research
Diagnostic Algorithm for Acute Kidney Injury (AKI)
The diagnostic algorithm for AKI in inpatients involves the use of various criteria and tools, including:
- Serum creatinine levels 2
- Urine output measurements 3, 2
- Electronic alert systems 4
- Machine learning algorithms 5
- Novel AKI biomarkers and techniques to measure glomerular filtration rate in real time 2
KDIGO Criteria
The Kidney Disease: Improving Global Outcomes (KDIGO) criteria are widely used to diagnose and classify AKI, and are based on changes in serum creatinine and urine output 3, 2. However, the KDIGO criteria have some limitations, including:
- Misdiagnosis of AKI can occur when using only the KDIGO criteria 2
- Neither creatinine nor urine output are renal-specific 2
Electronic Monitoring and Alert Systems
Electronic monitoring of urine output and alert systems can help identify AKI earlier and more accurately than traditional methods 3, 4. For example:
- Real-time consecutive urine output measurements can identify AKI patients earlier than serum creatinine criteria 3
- Electronic alert systems can facilitate early review and intervention for AKI 4
Machine Learning Algorithms
Machine learning algorithms, such as the NAVOY Acute Kidney Injury algorithm, can predict the onset of AKI with high accuracy using clinical variables and electronic health record data 5. These algorithms have the potential to strengthen patient monitoring and management, and improve patient outcomes.
Novel Diagnostic Tools
Novel diagnostic tools, including biomarkers and techniques to measure glomerular filtration rate in real time, are being developed and validated to diagnose AKI earlier and more accurately 2. These tools may offer improved diagnostic accuracy and earlier detection of AKI, enabling timely intervention and management.