Editorial
Definition and Classification of Kidney Diseases
Related Articles, p. 649 and 673
K DIGO (Kidney Disease: Improving Global Out-
comes) recently has completed development of
a series of new clinical practice guidelines that cover a
broad range of topics, including acute kidney injury
(AKI), glomerulonephritis, hypertension, lipids, ane-
mia, and chronic kidney disease (CKD).1 Guideline
development is “global,” but implementation is “local,”
so it is common for national and regional guideline
groups to issue commentaries on KDIGO guidelines.
This month’s issue of AJKD includes commentaries by
NKF-KDOQI (National Kidney Foundation–Kidney Dis-
ease Quality Outcome Initiative)2 and the CSN (Cana-
dian Society of Nephrology)3 on the KDIGO AKI
guideline published in 2012,4 and the ERBP (Euro-
pean Renal Best Practices) commentary was published
earlier in Nephrology Dialysis Transplantation.5 Of note,
the ERBP, KDOQI, and CSN commentaries differ in
their level of agreement with key aspects of the KDIGO
guideline’s definition and classification of AKI. In this
editorial, we review the rationale for the definitions and
classification of acute and chronic kidney diseases in the
guideline and offer some perspective about interpreting
these disagreements.
Kidney disease is defined as a heterogeneous group
of disorders affecting kidney structure and function. It
is recognized now that even mild abnormalities in
measures of kidney structure and function are associ-
ated with increased risk for developing complications
in other organ systems as well as mortality, all of
which occur far more frequently than kidney failure.
Duration of greater than 3 months is defined as
chronic, while duration of 3 months or fewer is termed
acute. AKI is defined as a subgroup of acute kidney
diseases and disorders (AKD) in which changes in
kidney function evolve within one week. There is a
complex relationship between AKI and CKD; AKI
can lead to CKD, and CKD increases the risk of AKI.6
The rationale for developing definitions and classi-
fications for kidney disease is based on the idea that
uniform terminology and explicit and objective crite-
ria can enhance communication and awareness, en-
able earlier detection and intervention, and ultimately
Address correspondence to Andrew S. Levey, MD, William B.
Schwartz Division of Nephrology, Tufts Medical Center, Box
391, 800 Washington St, Boston, MA 02111. E-mail: alevey@
tuftsmedicalcenter.org
© 2013 by the National Kidney Foundation, Inc.
0272-6386/$36.00
http://dx.doi.org/10.1053/j.ajkd.2013.03.003
686
improve outcomes. In principle, diseases can be de-
fined and classified according to many domains: struc-
ture, function, cause, duration, and outcomes (Fig 1).
Classification systems are in constant evolution to
accommodate the rapidly expanding information base
in clinical medicine and research. Kidney disease
differs from most other disorders of organ systems in
that it is often “silent”: there are few symptoms until
late in the course of disease, and even then the
symptoms are often nonspecific. Thus, there are few
kidney-specific “clinical events,” which results in a
reliance on laboratory measures to define the major
clinical syndromes. The duration of kidney disease
must be established either by documentation or infer-
ence of the duration of abnormalities in laboratory
measures.
The current definitions for CKD and AKI were
created by independent guideline development work
groups. They are based on the strong, graded, and
consistent relationship of laboratory measures with
important outcomes, independent of age, sex, race,
location, and comorbid conditions. Criteria include
measures of kidney damage (albuminuria, abnormali-
ties in the urine sediment, imaging, or biopsy) and
function (decreased glomerular filtration rate [GFR],
rising serum creatinine level, or decreased urine out-
put; Table 1). Both CKD and AKI are classified into
stages based in part on the severity in abnormalities in
these measures, and clinical practice guidelines fea-
ture a stage-based approach to evaluation and manage-
ment for both disorders. Identifying the cause of
kidney disease is not included in the definition of
either CKD or AKI. This omission was deliberate,
enabling the detection of kidney disease in epidemio-
logical studies and in many clinical settings in which
the cause of kidney disease is unknown or undocu-
mented. Identifying CKD and AKI by laboratory data
Function
Cause
•a
•b
•c
Structure
Outcome
Duration
Figure 1. Domains for the classification in kidney disease.
Am J Kidney Dis. 2013;61(5):686-688
Editorial

Table 1. Criteria for the Definitions of Kidney Diseases and Disorders

Functional Criteria Structural Criteria

AKI Increase in SCr by 50% within 7 d, or increase in SCr by 0.3 mg/dL within 2 d, or oliguria No criteria
CKD GFR ⬍60 mL/min for ⬎3 mo Kidney damage for ⬎3 mo
AKD AKI, or GFR ⬍60 mL/min/1.73 m2 for ⬍3 mo, or decrease in GFR by ⱖ35% or increase Kidney damage for ⬍3 mo
in SCr by ⬎50% for ⬍3 mo
NKD GFR ⱖ60 mL/min/1.73 m2, stable SCr No damage
Note: Criteria for AKI and CKD proposed by KDIGO guidelines, based on evidence and expert opinion. Criteria for AKD and NKD
proposed to harmonize the criteria for AKI and CKD. GFR may be assessed from measured or estimated GFR. Estimated GFR does
not reflect measured GFR in AKI as accurately as in CKD. Kidney damage assessed by pathology, urine or blood markers, imaging,
and—for CKD—presence of a kidney transplant. NKD indicates no functional or structural criteria according to the definitions for AKI,
AKD, or CKD. Clinical judgment is required for individual patient decision making regarding the extent of evaluation that is necessary to
assess kidney function and structure.
Abbreviations: AKD, acute kidney diseases and disorders; AKI, acute kidney injury or impairment; CKD, chronic kidney disease;
GFR, glomerular filtration rate; NKD, no known kidney disease or disorders; SCr, serum creatinine.

alone risks “overdiagnosis,”7 but it was the opinion of
the work groups that the risk of “underdiagnosis” was
potentially greater.
The original CKD guideline, which was published
by KDOQI in 2002, consolidated several lines of
evidence that recognize the utility of albuminuria as a
marker of kidney damage, decreased estimated GFR
as a means of staging CKD, and both as risk factors
for cardiovascular disease.8 Using these criteria for
the definition for CKD, epidemiological studies iden-
tified an average population prevalence of CKD of
10%-15% worldwide. These higher-than-anticipated
prevalence estimates initiated controversy about the
heterogeneity of prognosis within stages, the methods
used to characterize the kidney measures and the
terms used to describe the stages, and the limited
therapies available to treat early stages of kidney
disease. To address these controversies, KDIGO spon-
sored a conference in 2009 to critically evaluate data
on the prognosis of earlier stages of kidney disease
using large databases and uniform methods.9 The
findings showed more similarities than differences
among studies, leading to a consensus to retain the
KDOQI definition of CKD, but to modify the classifi-
cation to include cause of disease and albuminuria
stages. In response, KDIGO convened a guideline
work group, which concurred with recommendations
from the controversies conference and updated other
recommendations based on the data accumulated dur-
ing the past decade. The KDIGO CKD guideline10
was published in January 2013 and commentaries are
being prepared.
Similar to the process for CKD, the 2012 KDIGO
AKI guideline attempted to harmonize earlier defini-
tions for injury or impairment based on changes in
serum creatinine and urine output as measures of
kidney function.4 Markers of damage are not included
as criteria for defining AKI, although a number of
urinary biomarkers are under investigation as poten-
tial markers of kidney damage. The commentaries by
ERBP, CSN, and KDOQI emphasize broad areas of
agreement with the KDIGO guidelines, but there are
disagreements. In particular, the KDOQI group dis-
agreed with the inclusion of a 0.3-mg/dL rise in serum
creatinine as a criterion in the definition, and the
ERBP group disagreed with using a baseline serum
creatinine obtained prior to the acute illness or estimat-
ing the baseline serum creatinine as the expected
value for a baseline GFR of 75 mL/min/1.73 m2. All
groups found limitations in the staging system. While
some might characterize these disagreements as “con-
troversy,” we believe that they reflect differences in
emphasis about the interpretation of data rather than
disagreements about data. These issues were thor-
oughly discussed by the KDIGO work group and the
KDIGO Board of Directors as well as in the public
review of the draft guideline. In particular, we would
emphasize that AKI remains a clinical rather than a
laboratory diagnosis. As such, clinicians must exer-
cise their judgment in deciding whether a given pa-
tient who meets the criteria for the definition of AKI has
an important kidney disease or disorder, or conversely,
whether a patient who fails to meet the definition never-
theless has an important condition. By extension, it is
logical that the stages of AKI, which flow directly from
the definition, are also subject to clinical interpretation.
We agree with the commentary authors that clinicians
must not rely exclusively on laboratory data in diagnos-
ing or managing AKI, but should also consider cause
and a multitude of other clinical data. Indeed, a portion
of the guideline was devoted to clinical judgment pre-
cisely for this reason.
Although the conceptual models for AKI and CKD
are similar, there are important kidney diseases and
disorders which do not meet the specific criteria for
either AKI or CKD. Examples include structural dam-
age that lasts less than 3 months without decreased
GFR (eg, acute glomerulonephritis, childhood idio-

Am J Kidney Dis. 2013;61(5):686-688 687


pathic nephrotic syndrome, acute pyelonephritis, uni-
lateral obstruction) and a decline in GFR to ⬍60
mL/min/1.73 m2 for less than 3 months due to any
cause but without a sufficient rise in serum creatinine
to meet the criteria for AKI (eg, slow or small decline
in GFR in patients with mild reduction in baseline
GFR). To facilitate the identification of these condi-
tions, the AKI work group included a discussion of
the use of estimated GFR (eGFR) in acute kidney
disease and changes in eGFR that correspond to AKI
stages (a 1.5-, 2-, and 3-fold increase in serum creati-
nine corresponds to a 39%, 57%, and 74% decline in
eGFR, respectively). In addition, the AKI work group
suggested criteria for AKD that would be consistent
with definitions for AKI and CKD and recommended
that research studies refine these criteria (Table 1).
People who do not meet the criteria for AKD, CKD,
or AKI would be considered to have “no known
kidney disease.” We agree with the commentaries that
it would be premature to adopt these criteria into
clinical practice, but it seems obvious that the concept
of AKD would be an essential component of a compre-
hensive nomenclature for diseases of the kidneys.
In our view, the KDIGO definitions and classifica-
tions of AKI and CKD provide a roadmap. This
roadmap has now been validated by millions of data
points and can be relied on—but just as the wary
driver must not simply drive to the map, the clinician
must remain alert for all manner of obstacles and
choose a path that meets the clinical circumstances.
The KDIGO definitions and classifications, like all
clinical classifications, should be used with caution by
professionals who understand the various pitfalls that
may be encountered. We believe that all the commen-
taries have provided an important service by highlight-
ing these concerns. While not everyone will agree with
the KDIGO definitions and classifications of AKI and
CKD, we would encourage health care professionals not
to let these disagreements delay the implementation of
the guideline into clinical practice, research, and public
health. Moreover, consistent use of uniform definitions
and classifications will promote awareness and identifi-
cation of disease, aid clear communication, and foster
further clinical studies. Over time and using new data,
the definitions and classifications may be refined.
In conclusion, the KDIGO definitions and classifi-
cations for AKI and CKD represent the best currently
available consensus, are based on the solid evidence
and expert opinion, and provide a sound basis for
current practice, ongoing research, and health policy.
More research is necessary to determine whether
these classification systems improve patient out-
comes.
688
Levey, Levin, and Kellum
Andrew S. Levey, MD
Tufts Medical Center
Boston, Massachusetts
Adeera Levin, MD
University of British Columbia
Vancouver, Canada
John A. Kellum, MD
University of Pittsburgh
Pittsburgh, Pennsylvania
ACKNOWLEDGEMENTS
Dr Kellum was Co-Chair for the KDIGO Work Group on Acute
Kidney Injury. Dr Levin was Co-Chair for the KDIGO Work
Group on Evaluation and Management of Chronic Kidney Dis-
ease. Dr Levey was Chair for the KDOQI Work Group on Chronic
Kidney Disease: Classification, Evaluation and Stratification of
Risk.
Support: None.
Financial Disclosure: The authors declare that they have no
relevant financial interests.
REFERENCES
1. Kidney Disease: Improving Global Outcomes (KDIGO). Clini-
cal practice guidelines. http://www.kdigo.org/clinical_practice_
guidelines/index.php. Accessed February 14, 2013.
2. Palevsky PM, Liu KD, Brophy PD, et al. KDOQI US
commentary on the 2012 KDIGO clinical practice guideline for
acute kidney injury. Am J Kidney Dis. 2013;61:649-672.
3. James M, Bouchard J, Ho J, et al. Canadian Society of
Nephrology commentary on the 2012 KDIGO clinical practice
guideline for acute kidney injury. Am J Kidney Dis. 2013;
61:673-685.
4. Kidney Disease: Improving Global Outcomes (KDIGO) Work
Group. KDIGO clinical practice guideline for acute kidney injury.
Kidney Int Suppl 2012;2:1-138.
5. Fliser D, Laville M, Covic A et al. A European Renal Best
Practice (ERBP) position statement on the Kidney Disease Improv-
ing Global Outcomes (KDIGO) Clinical Practice Guidelines on
Acute Kidney Injury: part 1: definitions, conservative management
and contrast induced nephropathy. Nephrol Dial Transplant. 2012;
27:4263-4272.
6. Bedford M, Farmer C, Levin A, Ali T, Stevens P. Acute
kidney injury and CKD: chicken or egg? Am J Kidney Dis. 2012;
59:485-491.
7. Moynihan R, Doust J, Henry D. Preventing overdiagnosis:
how to stop harming the healthy. BMJ. 2012;344:e3502.
8. National Kidney Foundation. K/DOQI clinical practice guide-
lines for chronic kidney disease: evaluation, classification and
stratification. Am J Kidney Dis. 2002;39(suppl 1):S1-S266.
9. Levey AS, de Jong PE, Coresh J, et al. The definition,
classification and prognosis of chronic kidney disease: a KDIGO
controversies conference report. Kidney Int. 2011;80:17-28.
10. Kidney Disease: Improving Global Outcomes (KDIGO)
Work Group. KDIGO clinical practice guideline for the evaluation
and management of chronic kidney disease. Kidney Int Suppl.
2013;3:1-150.
Am J Kidney Dis. 2013;61(5):686-688