Redefining the Laboratory Approach to Acute Kidney Injury Investigation

The diagnosis of acute kidney injury (AKI) has significantly evolved over the past 20 years. Historically, blood creatinine (Cr) provided the mainstay of clinical laboratory assessment to identify AKI patients. A marker of renal function, Cr is known to represent a crude estimation of renal functional and structural integrity. Up to 50% of nephron mass may be lost without a significant elevation in Cr. The limitations of Cr have led to a decades-long search for improved methods to evaluate patients with suspected AKI (1).

Driven by evidence indicating care of patients with AKI was inadequate and risk factors were poorly recognized, The Kidney Diseases Improving Global Outcomes (KDIGO) working group established guidelines to define and stage AKI. The initial expert consensus guideline defined AKI by an increase in Cr by ≥0.3 mg/dL within 48 hours or an increase in Cr ≥1.5 times baseline within the previous 7 days or urine output ≤0.5 mL/kg/h for 6 hours (2). Since the initial report, application of the KDIGO guidelines has been demonstrated to improve clinical outcomes in a variety of clinical scenarios (3). The KDIGO AKI definitions and staging criteria represent a critical milestone in AKI patient management. Creation of standard criteria enabled investigators to compare patients across studies and provided clinicians a methodology to guide care. Along with progress, gaps have been identified; there is a high rate of false positive AKI diagnosis among patients with chronic kidney disease (CKD) and a high rate of false negative AKI diagnosis among patients with baseline Cr values less than 1.0 mg/dL. Implementation of urine output for AKI diagnosis outside of intensive care units creates additional challenges.

Once a patient is diagnosed with AKI, clinicians work to determine etiology. Although in the majority of cases, a prerenal etiology may be clear from chart review, this is not always the case. A battery of laboratory tests is commonly utilized to determine whether a patient has prerenal or intrinsic renal AKI. These include fractional excretion of urea (FEUr) and sodium (FENa), among others .

In this issue of The Journal of Applied Laboratory Medicine, El-Khoury et al. provide expert guidance, based on literature review, on current best practice for laboratory investigation of AKI, incorporating traditional biomarkers, emerging biomarkers, automated AKI alerts, and analytic and biologic variation (4). The multidisciplinary group of nephrologists and laboratory scientists addressed a number of recent developments since the initial publication of the KDIGO AKI guidelines. The modern laboratory will find the approach to be relevant and will enable individual laboratories to customize diagnostic criteria for AKI based on their individual laboratory assay analytic performance.

The authors provide a detailed quantitative approach to measurement of Cr. They propose incorporating analytic and biologic variation into a revised definition of AKI. Intraindividual biologic variation of Cr is ∼4.5% and intralaboratory analytical variation is 1.0%–3.0% for enzymatic or Jaffe assays, respectively (4). For a Cr assay to meet minimum precision criteria, intralaboratory variability should be less than 75% of the intrabiologic variation, or 3.4% (4). To address the issue of biologic and assay variability, the authors propose the 20/20 rule; changes in Cr by 0.2 mg/dL or 20%, whichever is greater, are interpreted as within analytical and biological variation. Patients with Cr <1.0 mg/dL require a 0.2 mg/dL increase from baseline and patients with Cr >1.0 mg/dL require a 20% increase from baseline to reach AKI (4). This proposal is aimed at improving Cr sensitivity and specificity for AKI diagnosis. As with KDIGO, the current recommendations rely on an accurate baseline Cr definition.

Defining a patient’s baseline Cr has been explored by a number of investigators and is thoroughly discussed by El-Khoury et al. (4, 5). The authors conclude that there is insufficient evidence to recommend one approach over another. However, some evidence suggests that the KDIGO recommendation to use the lowest value in the last 12 months is linked to tubular injury pathology and to improved outcomes when implemented as part of an automated AKI alert (6, 7). Utilizing the KDIGO definition or, in the absence of such a value, a calculated baseline, represent a straightforward solution for laboratories to standardize this important parameter. Further studies are needed to better understand how variance in baseline definitions impact clinical outcomes and tubular injury histopathology.

Redefining AKI based on the 20/20 rule opens an opportunity to implement revised electronic AKI alerts. Previous implementations of automated AKI alerts using KDIGO criteria have had mixed results (8). This may be related to ineffective intervention strategies or a lack of clinician education regarding protocols for managing AKI when an alert is triggered. Alternatively, the KDIGO definition of AKI may miss AKI patients with low baseline Cr and overestimate AKI in the CKD population. Identifying AKI patients using the suggested 20/20 rule addresses the sensitivity and specificity issues. Future studies evaluating the 20/20 cutoffs in an automated alert system will be important to assess utility in clinical practice.

It has been common practice for clinicians to request urine sodium and urea for the purposes of distinguishing prerenal from intrinsic renal AKI. The authors provide a thorough evaluation of the ability of these widely utilized biomarkers to perform as intended. Notably, the FENa shows poor performance—AUC 0.59—for discriminating prerenal from intrinsic renal AKI in patients with sepsis due to variations in volume status and diet (9). Similarly, FEUr may vary significantly depending on the amount of dietary protein or variance in liver function. As a result, the authors only recommend using FENa and FEUr as supportive tests requiring careful interpretation within clinical context. They are not indicated for routine general use, with the exception of patients who have hepatorenal syndrome.

Urine microscopy is frequently performed to evaluate patients with AKI. This test is limited by a high degree of interobserver variability and a lack of automated, standardized analytic tools. This area is open to improvements in clinical laboratory testing. Investigation and development of novel image analysis techniques to move urine microscopy from a qualitative to a quantitative test should be considered.

Over the last 15 years, the field of AKI biomarkers has rapidly expanded but has yet to lead to widespread adoption of new biomarkers for clinical use (10). Ideal assessment of AKI requires accurate measurement of renal function and loss of renal structural integrity. A single biomarker is unlikely to meet both characteristics. An ideal AKI biomarker to assess tissue integrity would display similar characteristics to troponin: highly abundant and specific to the tissue of interest, rapidly measured in a standardized way, and correlating with the degree of tissue damage. Combination testing using the functional biomarker, cystatin C, and a damage biomarker, neutrophil gelatinase-associated lipocalin, has been shown to be superior to Cr in predicting AKI severity among children undergoing cardiopulmonary bypass surgery (11). The Food and Drug Administration (FDA) approved measurement of urinary insulin-like growth factor binding protein 7 (IGFBP7) and tissue inhibitor of metalloproteinases 2 (TIMP2) for use in the USA for moderate and severe AKI risk assessment (12). Since IGFBP7: TIMP2 is FDA approved, the authors focused their analysis on clinical implementation of this assay.

The urinary IGFBP7: TIMP2 assay showed impressive performance in the Sapphire study for identifying critically ill adult patients who are at risk of developing AKI (13). The established threshold for AKI risk, however, shows significant overlap with urine IGFBP7: TIMP2 in adults without AKI. As a result, the assay sensitivity is high, and specificity is low. There is a lack of information regarding biologic variability of IGFBP7: TIMP2 and assay performance outside of the critically ill population. The authors argue that the assay shows variation based on timing of collection relative to an AKI event. This particular critique would be applicable to virtually any AKI assay. All laboratory tests are interpreted in clinical context with an understanding of the biologic time course for a given biomarker. While the IGFBP7:TIMP2 test is ideal in the immediate period following AKI, and thereby able to open a therapeutic opportunity, it will likely need to be implemented as a panel of AKI biomarkers in order to identify a specific patient’s time course. When the inciting event is known, such as cardiac surgery, urine IGFBP7:TIMP2 may assist with AKI risk assessment. There are insufficient data, however, to recommend urine IGFBP7:TIMP2 in low-risk populations or for routine monitoring (4).

The current expert recommendations by El-Khoury et al. build on the success of KDIGO by providing laboratorians and clinicians with standard approaches to defining and investigating AKI. There are gaps in our current methods; particularly urine microscopy and traditional tests to define prerenal versus intrinsic renal causes. Advances in digital pathology and computer vision represent an opportunity for future work aimed to move urine microscopy from a qualitative to a quantitative test. Ultimately, continued revisions of AKI diagnosis using Cr will continue to suffer from the inherent shortcomings of Cr as an AKI biomarker. If the field is to move toward a more precise approach to AKI diagnosis, clinical adoption of more specific and sensitive AKI biomarkers will be required.

 

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.

Authors’ Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest: Employment or Leadership: None declared. Consultant or Advisory Role: None declared. Stock Ownership: None declared. Honoraria: None declared. Research Funding: J.P. Gaut, NIH. Expert Testimony: None declared. Patents: J.P. Gaut is listed as a coinventor on a US patent for acute kidney injury biomarkers.

Other Remuneration: J.P. Gaut may receive royalty income based on whole slide image analysis technology developed in part by J.P. Gaut and licensed by Washington University to PlatformSTL. This technology is not related to the content of the editorial.