From the Chair: Becoming Precise in Diagnosing Acute Kidney Injury

Ajay K. Singh,
Brigham and Women’s
Hospital and Harvard
Medical School

From the time that I trained in nephrology more than 30 years ago, I have been teaching students and fellows the importance of framing the differential diagnosis of acute kidney injury (AKI) along anatomical lines: prerenal, renal, and postrenal, and contextualizing based on the presence or absence of oliguria and on the tempo of the disease process. In the past decade, however, the definition and classification of AKI has moved toward using serum creatinine-based criteria, such as the Risk, Injury, Failure, Loss and End-Stage Kidney Disease (RIFLE), The Acute Kidney Injury Network (AKIN), and Kidney Disease Improving Global Outcomes (KDIGO) guidelines. Are these criteria relevant clinically?

In an interesting Comment in the Lancet1, Jonathan Barasch and colleagues point to several limitations of a creatinine-based approach. They point out: “The uniform application of serum creatinine ‘stages,’ in lieu of a primary etiologic or anatomic diagnosis, (i) provides inadequate quantitative assessment of excretory dysfunction, and (ii) obfuscates the important distinctions among fundamentally different etiologies that raise serum creatinine and motivate personalized therapy.” They further make the point that “’AKI stages’ not only poorly describe the extent of defective excretory function, but they are often at variance with kidney pathology and physiology.”

In reply to the Barasch Lancet article, Kellum and Lamiere rebut what they call nostalgia “for a time when acute renal dysfunction was evaluated according to the classification of prerenal, intrarenal, and postrenal causes,” pointing out that several forms of AKI include intrarenal and extrarenal etiologies, for example sepsis, cardiac surgery, liver disease, and trauma.3 This seems to sidestep the issue. As nephrologists we have long appreciated that some forms of AKI have overlapping prerenal and intrarenal causes. We can handle this moderate degree of complexity but the question is: Does it justify abandoning a long-established framework to the diagnosis and management of AKI? Barasch and colleagues think not.

Kellum and Lamiere also point out that the KDIGO guidelines do not substitute for the importance of clinical judgment. Of course, most nephrologists understand this. But the point that Barasch and colleagues make is that a stage-based approach to AKI is intrinsically flawed because it is based on serum creatinine as a biomarker. The limitations of serum creatinine are well known. Rather, Barasch et al. point to emerging advances in kidney transcriptomics and urinary proteomics that overlaid on etiologic and anatomic diagnoses will be more clinically valuable than a stage based approach.

Pickering et al.2 and Lamiere3 have pointed out that RIFLE, AKIN, and KDIGO have key differences that might create confusion and the possibility of AKI misclassification. Indeed, RIFLE defines AKI by a change from baseline in the serum creatinine or estimated glomerular filtration rate (eGFR), and incorporates urine output over a specified time period. In contrast, AKIN defines AKI based on the RIFLE criteria but incorporates an absolute change in serum creatinine of ≥0.3 mg/dL but without eGFR criteria, and includes 48 hours as a time constraint. On the other hand, the KDIGO guidelines retain the AKIN staging criteria but allow a time frame of 7 days for a 50% increase in serum creatinine.

The door seems to be opening to a much more personalized genetic and proteomic approach. This would be most welcome, since our current ways of managing AKI remain rather rudimentary.

So the bottom line for me is that I will continue to teach fellows and students at the Brigham and Women’s Hospital where I practice to frame AKI both anatomically and etiologically. I will make them aware of the importance of the KDIGO guidelines and encourage them to apply them where relevant.


  1. Barasch J, Zager R, Bonventre JV. Acute kidney injury: A problem of definition Lancet. 2017;389(10071): 779–781.
  2. Pickering JW, Endre ZH. GFR shot by RIFLE: Errors in staging acute kidney injury. Lancet. 2009;373(9672):1318-1319.
  3. Kellum JA, Lameire N. The definition of acute kidney injury, Lancet. 2018; 391(10117):202-203.