The n e w e ng l a n d j o u r na l of
Review Article
From the Division of Nephrology, Tufts
Medical Center, Boston (A.S.L., L.A.I.); and
the Division of Precision Medicine, Depart-
ment of Medicine, New York University,
New York (M.E.G.). Dr. Levey can be con-
tacted at ­alevey@​­tuftsmedicalcenter​.­org
or at the Division of Nephrology, Tufts
Medical Center, 800 Washington St., Box
391, Boston, MA 02111.
This article was updated on June 2, 2022,
at NEJM.org.
N Engl J Med 2022;386:2120-8.
DOI: 10.1056/NEJMra2201153
Copyright © 2022 Massachusetts Medical Society.
CME
at NEJM.org
2120
Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
m e dic i n e
Julie R. Ingelfinger, M.D., Editor
Uses of GFR and Albuminuria Level
in Acute and Chronic Kidney Disease
Andrew S. Levey, M.D., Morgan E. Grams, M.D., Ph.D., M.H.S.,
and Lesley A. Inker, M.D.​​
K
idney disease is common in adults, and testing for kidney dis-
ease is part of routine clinical practice for patients with acute or chronic
illness. The initial evaluation includes determination of the glomerular
filtration rate (GFR), estimated on the basis of the serum creatinine level (eGFRcr),
and the level of albuminuria, assessed on the basis of the urinary albumin-to-
creatinine ratio. These tests are inexpensive, are widely available in clinical labo-
ratories, and enable early detection of most kidney diseases. Calculation of the
eGFR from the serum cystatin C level alone or with the serum creatinine level
(eGFRcys and eGFRcr-cys, respectively) is a recommended confirmatory test but is
less widely available, particularly in countries outside the United States and west-
ern Europe. Rigorous evaluation of GFR and albuminuria has enabled advances in
clinical practice, research, and public health, but greater use of these measures is
needed to improve health outcomes related to kidney disease.
It has been 20 years since the first clinical practice guideline on kidney disease
recommended the use of the GFR and albuminuria for defining and classifying
kidney disease.1 In this review, we briefly discuss the conceptual model and clini-
cal approach to detection, evaluation, and management of acute kidney disease
(AKD) and chronic kidney disease (CKD), using internationally recommended
nomenclature.2,3 We then focus on advances in clinical evaluation of the GFR and
albuminuria and application of these measures as criteria for the definitions of
AKD and CKD, risk factors for cardiovascular disease, eligibility criteria and end
points for clinical trials of kidney disease progression, and risk prediction instru-
ments for clinical practice (Table 1).
Cl inic a l A pproach t o K idne y Dise a se
Current guidelines define kidney disease as heterogeneous disorders characterized
by abnormalities in the function or structure of the kidney, with implications for
health (Fig. 1).4-6 Functional abnormalities are related to a decreased GFR, where-
as structural abnormalities are inferred from markers of kidney damage, including
increased albuminuria and abnormalities in the urine sediment and imaging. New
biomarkers of kidney function and structure are under investigation.7 AKD and
CKD are differentiated by their duration: less than 3 months for AKD and 3 months
or longer for CKD. Acute kidney injury (AKI) is defined as a subset of AKD, with
an onset within 7 days before presentation. AKD with or without AKI can be super-
imposed on preexisting CKD.8 Kidney disease is classified according to the cause
and severity (stages) of GFR and albuminuria (known as cause–GFR–albuminuria
classification, or CGA staging) because of the importance of each of these factors
in prognosis and treatment.
n engl j med 386;22  nejm.org  June 2, 2022
The New England Journal of Medicine
 Uses of GFR and Albuminuria in Kidney Disease

Table 1. Clinical Uses of the Glomerular Filtration Rate (GFR) and the Level of Albuminuria.*

Clinical Use GFR Albuminuria

Clinical significance in the general Overall index of kidney function: decreased GFR re- Marker of kidney damage: increased albumin-
population flects decreased nephron number or decreased uria reflects impairment of permselectivity
single-nephron GFR barrier of glomerular capillaries to macro-
molecules†
Normal range in the general Mean value is approximately 125 ml/min/1.73 m2 Mean AER is approximately 5 mg/day in young
population in young adults, with a wide range; lower mean adults, equivalent to ACR of about 5 mg/g,
values at older age because of aging and kidney with a wide range; higher mean ACR at older
disease age due to higher AER and lower creatinine
excretion rate
Detection of kidney disease
AKI Urine output, increase in serum creatinine from NA
baseline
AKD or CKD Initial test: eGFRcr Initial test: ACR in untimed spot urine sample
Confirmatory tests: eGFRcys, eGFRcr-cys, mGFR, (early-morning sample preferred)
mClcr Confirmatory test: AER in timed urine sample
Definition of kidney disease
AKI Serum creatinine increased by 0.3 mg/dl in 48 hr or NA
by 50% in 7 days, or oliguria for >6 hr
AKD AKI or GFR <60 ml/min/1.73 m2 or GFR decreased ACR >30 mg/g for <3 mo
by 35% for <3 mo
CKD GFR <60 ml/min/1.73 m2 for ≥3 mo ACR >30 mg/g for ≥3 mo
Risk of adverse outcomes
AKI Increased risk of CKD, KFRT, death NA
AKD Increased risk of CKD, KFRT, death Insufficient data
CKD Increased risk of AKI, KFRT, CVD, death‡ Increased risk of AKI, KFRT, CVD, death‡
Clinical trials of CKD progression Predictive biomarker for eligibility Predictive biomarker for eligibility
Clinical end point: kidney failure (GFR <15 ml/ Surrogate end point: mean reduction in ACR
min/1.73 m2 or KFRT) >30%
Surrogate end points: 57%, 40%, 30% declines
in GFR or mean reduction in GFR slope
>0.5–1.0 ml/min/1.73 m2 per yr
Risk-prediction instruments for CKD CKD onset, CKD progression, KFRT, CVD, death CKD onset, CKD progression, KFRT, CVD, death
(absolute risk) in general popu-
lation and patients with CKD

* For the urinary albumin-to-creatinine ratio (ACR), albumin is measured in milligrams and creatinine in grams. AER denotes albumin excre-
tion rate, AKD acute kidney disease, AKI acute kidney injury, CKD chronic kidney disease, CVD cardiovascular disease, eGFRcr estimated
GFR based on creatinine, eGFRcr-cys estimated GFR based on creatinine and cystatin C, eGFRcys estimated GFR based on cystatin C, KFRT
kidney failure requiring long-term replacement therapy, mClcr measured creatinine clearance, mGFR measured GFR, and NA not applicable.
† Other markers of kidney damage in AKD and CKD include urine sediment abnormalities (hematuria and pyuria), imaging abnormalities,
pathological abnormalities, and electrolyte disorders due to altered tubular function.
‡ Adverse outcomes of CVD include death, coronary heart disease, stroke, heart failure, peripheral arterial disease, and amputation of any
lower extremity.

There are numerous risk factors for and
causes of AKD and CKD (see the text and Tables
S1 and S2 in the Supplementary Appendix, avail-
able with the full text of this article at NEJM
.org). Older age, diabetes, hypertension, and
obesity are among the most common risk fac-
tors. Regardless of the cause of disease, adverse
outcomes include progression to kidney failure,
complications in other organs and tissues, and
death. Kidney failure generally causes symptoms
and may require short-term or long-term kidney-
replacement therapy (dialysis or transplanta-
tion). Kidney failure requiring long-term kidney-
replacement therapy is also known as end-stage
kidney disease. Complications may involve virtu-
ally all organs and tissues, but causal mecha-
nisms are incompletely understood (Fig. 1). Kid-
ney failure and complications associated with

n engl j med 386;22  nejm.org  June 2, 2022 2121

The New England Journal of Medicine
Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Complications Related to Complications Related
Increased Albuminuria to Decreased GFR
Hypoalbuminemia Drug toxicity
Hyperlipidemia Acid–base or electrolyte
Deep-vein thrombosis complications
Pulmonary embolism Metabolic or hormonal
complications
Neuropathy
Complications Related to
Both Decreased GFR and
Increased Albuminuria
Cardiovascular disease
Fluid overload
Anemia
Malnutrition
Infections
Cognitive impairment
Frailty

Increased risk of
kidney disease Albuminuria as a Decreased GFR Kidney failure
Older age eGFRcr GFR <15 ml/
marker of damage
Normal Diabetes mellitus Death
Increased ACR eGFRcr-cys min/1.73 m2
High blood or dialysis
pressure
Obesity
AKI: onset within 7 days
AKD: duration <3 mo; CKD: duration ≥3 mo
Populations for application of risk-prediction instruments

Figure 1. Use of the Glomerular Filtration Rate (GFR) and Level of Albuminuria to Ascertain the Development, Progression,
and Complications of Acute and Chronic Kidney Disease.
The white rectangle lists complications associated with kidney disease; such complications refer to nonkidney outcomes that occur more
frequently in persons with than in those without kidney disease. The white squares indicate the antecedents and stages of kidney dis-
ease. The squares also list examples of risk factors for kidney disease and tests for evaluation of the level of albuminuria and the GFR.
Arrows pointing to the right represent the development and progression of kidney disease, and arrows pointing to the left represent re-
mission. Death from kidney disease may be due to kidney failure or complications associated with kidney disease. The rectangle with the
darkest shading indicates stages of acute kidney injury (AKI), the rectangle with medium shading indicates stages of acute kidney disease
(AKD) and chronic kidney disease (CKD), and the rectangle with the lightest shading indicates populations (e.g., persons with an increased
risk of kidney disease) that are appropriate for the application of risk-prediction instruments. ACR denotes albumin-to-creatinine ratio,
eGFRcr estimated GFR based on the creatinine level, and eGFRcr-cys estimated GFR based on the creatinine and cystatin C levels.

kidney disease (especially cardiovascular disease
[CVD]) contribute to decreased quality of life,
higher costs of care, and increased mortality,
especially among older adults.
The general clinical approach to AKD (with
or without AKI) and CKD consists of six steps
(Box S1): first, recognition of patients who are at
increased risk for kidney disease; second, testing
of patients at increased risk, even in the absence
of symptoms; third, detection with the use of
guideline-recommended criteria; fourth, evalua-
tion of cause and stages (cause–GFR–albumin-
uria classification); fifth, assessment of progno-
sis (risk) for kidney disease progression and
complications; and sixth, implementation of risk-
based therapy, including cause-specific therapy
and stage-based nonspecific therapy. Treatment
is designed to slow progression and reduce com-
plications for patients in whom the anticipated
benefits of treatment (risk reduction) outweigh
the anticipated harms.
Cl inic a l E va luat ion of GFR a nd
A l buminur i a
GFR and albuminuria reflect the glomerular
contributions to the excretory function of the
kidneys and are the most well-characterized
measures of kidney disease. GFR, the product of
the number of nephrons and the average single-
nephron GFR, is generally considered the best
overall index of kidney function in health and
disease.9 The nephron number declines with age
and kidney disease.10,11 Physiologically, the con-
sequent decline in GFR can be counteracted by
an increase in single-nephron GFR (single-neph-
ron hyperfiltration), which can blunt the decline
in the overall GFR, but over time, this may result

2122 n engl j med 386;22  nejm.org  June 2, 2022

The New England Journal of Medicine

Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
 Uses of GFR and Albuminuria in Kidney Disease
in hemodynamic injury to the remaining neph-
rons and accelerated progression of kidney disease.
The true GFR cannot be measured directly in
humans. GFR can be assessed from the clear-
ance or plasma (or serum) concentrations of
filtration markers (low-molecular-weight sub-
stances that are eliminated primarily by glo-
merular filtration). For evaluation of the GFR,
current guidelines recommend using GFR-esti-
mating equations, with eGFRcr as the initial
test, and confirmation with eGFRcys, eGFRcr-cys,
or clearance measurements, as appropriate for
the clinical setting (see the text in the Supple-
mentary Appendix).5 GFR can be assessed more
accurately with the use of estimating equations
than by the concentration of the filtration marker
alone. The equations developed by the Chronic
Kidney Disease Epidemiology Collaboration are
recommended for use in North America, Europe,
and Australia. Alternative equations for eGFRcr
are used in some regions of Asia; there are lim-
ited studies from South America and Africa.12
Furthermore, recent recommendations encour-
age the use of serum cystatin C levels in estimat-
ing the GFR because cystatin C is not affected by
race or world region and appears to enable a
more accurate eGFR determination.13-15 Assays
for cystatin C are more expensive than those for
creatinine, and use of cystatin C is currently
limited by reimbursement policies, but there are
opportunities to reduce its costs and advocate
for changes in policy.
The accuracy of the eGFR is affected by non-
GFR determinants of the filtration markers,
which are better understood for creatinine than
for cystatin C (Table S3).16 Creatinine is affected
by muscle mass, diet, and drugs that interfere
with tubular secretion (e.g., trimethoprim). As
compared with creatinine, cystatin C is less af-
fected by age, sex, race, and world region but is
more affected by obesity, smoking, inflamma-
tion, and alterations in thyroid and glucocorti-
coid hormones. The accuracy of eGFRcr-cys is
greater than that of either eGFRcr or eGFRcys
because the use of both creatinine and cystatin C
reduces the error due to the non-GFR determinants
of the two filtration markers. Measured GFR
(mGFR), determined on the basis of urinary or
plasma clearance of an exogenous filtration
marker, is more accurate than eGFR; however,
such methods are typically performed only in
n engl j med 386;22  nejm.org  June 2, 2022
The New England Journal of Medicine
Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
referral centers or as part of research studies.
Urinary clearance of endogenous creatinine, gener-
ally assessed from a single serum measurement
and a 24-hour urine collection for measurement of
the creatinine excretion rate, is readily tested but
is less accurate than mGFR and may be less ac-
curate than eGFR.
Increased albuminuria, reflecting impairment
of the permselective barrier function of the glo-
merular capillary wall to macromolecules, is a
marker of kidney damage. Increased albumin-
uria is seen in early stages of kidney disease due
to diabetes, other glomerular diseases, or hyper-
tension and in later stages of almost all causes
of kidney disease.9 Numerous pathologic pro-
cesses, such as inflammation, infiltration, or fi-
brosis, can cause albuminuria. Irrespective of the
cause, albuminuria and the presence of other
macromolecules in the tubular fluid may incite
tubular damage and accelerate the progression
of kidney disease.
Albuminuria can be assessed by means of
albumin-specific assays or as a component in
total urinary protein, with the use of quantita-
tive or semiquantitative methods, in either a
timed urine collection or an untimed spot urine
sample (Table S4). Current guidelines for albu-
minuria evaluation recommend determining the
albumin-to-creatinine ratio in a spot urine sam-
ple as the initial test (an early-morning sample
is preferred), with the result confirmed by as-
sessing the albumin excretion rate in a timed
urine collection, as appropriate for the clinical
setting (see the text in the Supplementary Ap-
pendix).5 Assays for urinary albumin may be
more expensive than assays for urinary total
protein or dipstick testing, and use of the albu-
min-to-creatinine ratio is limited by reimburse-
ment policies in some countries. Equations have
been developed to estimate the urinary albumin-
to-creatinine ratio from the protein-to-creatinine
ratio and dipstick testing of protein, but accu-
racy is limited at low protein-to-creatinine ratios
and low dipstick protein values.17,18
Cl inic a l A ppl ic at ion of GFR
a nd A l buminur i a
Definitions and Burden of AKD and CKD
Either a decreased GFR (<60 ml per minute per
1.73 m2 of body-surface area) or increased albu-
2123
 Table 2. Predictive Instruments Using GFR and Albuminuria.*

2124
Interval from Recommended
Risk Prediction No. of Predicted Risk in Risk Threshold
Instrument and Target Population Outcome to Outcome Hazard Ratio† Predictors Target Population‡ for Clinical Actions§
GFR ACR 1st decile 10th decile
years percent percent
KFRE: GFR <60 ml/min/1.73 m2 KFRT 2 29 2.6 4 <0.1 93 >10: Multidisciplinary care
>20–40: Dialysis access and trans-
plantation evaluation
KFRT 5 29 2.6 4 <0.1 100 >5: Nephrology referral
CKD with Severely Decreased GFR: KFRT 2 4.1 3.9 9 0.8 42 Same as for KFRE; high risk of CVD
GFR <30 ml/min/1.73 m2 CVD 2 0.9 1.2 9 3.8 34 and death influences overall
Death 2 1.3 1.0 9 2.6 36 management
KFRT Risk Tool for Kidney Donor Can­ KFRT Lifetime 2.3 2.6 10 0.1 6.0 <1–2: Acceptable donor candidate
didates: GFR ≥60 ml/min/
1.73 m2 and ACR <300 mg/g
The

CKD Progression Calculator:

GFR ≥60 ml/min/1.73 m2, no DM 40% GFR decline 3 1 2.4 12 0.2 7.8 >1–4: Use medications to slow
GFR ≥60 ml/min/1.73 m2, DM 40% GFR decline 3 1.4 2.3 12 0.4 18 kidney-disease progression,
GFR <60 ml/min/1.73 m2, no DM 40% GFR decline 3 2.7 2.3 12 1.4 42 more frequent monitoring
GFR <60 ml/min/1.73 m2, DM 40% GFR decline 3 1.6 2.6 12 1.2 65 >5: Consider multiple medications
to slow disease progression
Incident GFR <60 Calculator:
GFR ≥60 ml/min/1.73 m2, no DM GFR <60 ml/min/1.73 m2 5 13 1.4 9 0.1 46 >5: Measure albuminuria if not
GFR ≥60 ml/min/1.73 m2, DM GFR <60 ml/min/1.73 m2 5 6.9 1.4 11 2.1 68 done, use medications to slow
disease progression
SCORE plus Kidney Variables Calculator: Death from CVD 10 3.0 1.6 7 0.6 27 >5: Improves risk prediction
general population free of CVD beyond SCORE¶
n e w e ng l a n d j o u r na l

PCE plus Kidney Variables Calculator: ASCVD event 10 1.7 1.3 11 0.3 54 Same as for SCORE plus Kidney

The New England Journal of Medicine

of

general population free of CVD Variables Calculator

* Shown are instruments developed and validated in large, generalizable settings by the Chronic Kidney Disease Prognosis Consortium (https://ckdpcrisk​.o ­ rg/​­). Web links to the in-
struments are as follows: KFRE (https://ckdpcrisk​.­org/​­kidneyfailurerisk/​­), CKD with Severely Decreased GFR (https://ckdpcrisk​.­org/​­lowgfrevents/​­), KFRT Risk Tool for Kidney Donor

n engl j med 386;22  nejm.org  June 2, 2022

Candidates (https://ckdpcrisk​.­org/​­esrdrisk/​­), CKD Progression Calculator (https://ckdpcrisk​.­org/​­gfrdecline40/​­), Incident GFR <60 Calculator (https://ckdpcrisk​.­org/​­ckdrisk/​­), SCORE
plus Kidney Variables Calculator (https://ckdpcrisk​.­org/​­ckdpatchscore/​­), and PCE plus Kidney Variables Calculator (https://ckdpcrisk​.­org/​­ckdpatchpce/​­). The cause of CKD is generally

Copyright © 2022 Massachusetts Medical Society. All rights reserved.

m e dic i n e

not known and is not included in these instruments. Other instruments are available for use in CKD with specific identified causes. ASCVD denotes atherosclerotic cardiovascular dis-
ease, DM diabetes mellitus, KFRE Kidney Failure Risk Equation, PCE Pooled Cohort Equations, and SCORE Systematic Coronary Risk Evaluation.
† Hazard ratios are for a GFR of 65 versus 95 ml/min/1.73 m2 for patients with a GFR of 60 ml/min/1.73 m2 or higher, 25 versus 55 ml/min/1.73 m2 for a GFR of less than 60 ml/
min/1.73 m2, and 15 versus 30 ml/min/1.73 m2 for a GFR of less than 30 ml/min/1.73 m2 and a urinary ACR (with albumin measured in milligrams and creatinine in grams) that is in-
creased by a factor of 8 (e.g., 320 vs. 40 mg/g or 40 vs. 5 mg/g). Hazard ratios are rounded and, in some cases, are averages across nonlinear associations.
‡ The risks are based on data from multiple U.S. health care systems. Deciles are categorized by predicted risk for each outcome computed from the target population in OptumLabs
Data Warehouse (OLDW) cohorts. The OLDW is a longitudinal, real-world data asset with deidentified administrative claims and electronic health record data. The 1st decile represents
the lowest 10%; the 10th decile represents the highest 10%. The range between the 1st and 10th deciles allows an evaluation of the range of predictions with respect to risk thresholds.
For example, for KFRT, the risk at 2 years ranges from <0.1 to 93%; thresholds for clinical actions at 10 to 40% are well within this range.

Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
§ Risk thresholds are based on expert opinion regarding the implementation of risk-based, guideline-directed therapy. The thresholds are likely to vary among clinical settings (e.g., high-
resource settings vs. low-resource settings) and to change as experience with guideline implementation accumulates.
¶ Examples of other CVD risk-reduction strategies are as follows: for a risk threshold of 5 to 7.5%, discuss risk and consider the use of a moderate-intensity statin; for a risk threshold of
greater than 7.5 to 20%, use a moderate-intensity statin; and for a risk threshold of >20%, use a high-intensity statin.
 Uses of GFR and Albuminuria in Kidney Disease
minuria (albumin-to-creatinine ratio >30 mg per
gram, with albumin measured in milligrams
and creatinine in grams) constitutes the clinical
criterion for the definitions of AKD and CKD,
whereas the only criterion for AKI is a decreased
GFR (indicated by oliguria or an increase in the
serum creatinine level from baseline) (Table S5).
The thresholds for the disease definitions do not
vary according to age, clinical characteristics, or
coexisting conditions.
A lower GFR and higher level of albuminuria
are associated with greater risks of kidney fail-
ure that requires replacement therapy (dialysis or
transplantation), a broad spectrum of treatable
complications (e.g., CVD), and death (Table 1
and Figs. S1 and S2).19-22 These risk relationships
provide the rationale for the GFR and albumin-
uria thresholds used for disease definitions, as
well as for the GFR and albuminuria categories
used for staging of disease severity and guide-
line-directed care (Table S5). In CKD, a lower
GFR and higher level of albuminuria are associ-
ated with an increased risk of adverse outcomes,
regardless of patient age or the presence or ab-
sence of diabetes, hypertension, or obesity; thus,
a decreased GFR or increased albuminuria can-
not be considered normal in any of these sub-
groups.23-26 A decreased GFR is an indication for
adjustment of drug dosages; testing for electro-
lyte and acid–base disorders, anemia, and meta-
bolic bone disease; and planning for initiation of
kidney-replacement therapy. Increased albumin-
uria is an indication for treatment with renin–
angiotensin–aldosterone inhibitors. Both are
indications for strict blood-pressure control and
sodium–glucose cotransporter 2 inhibition to
slow the progression of kidney disease and re-
duce the risk of CVD.
The aging population and the obesity epi-
demic are leading to a growing burden of AKD
and CKD globally. In Alberta, Canada, the 1-year
incidence is 2.5% for AKD (0.75% for AKI and
1.75% for AKD without AKI).27 Each disorder
has been associated with a higher risk of adverse
outcomes and death, as compared with no kid-
ney disease. In the United States, the reported
prevalence of CKD among adults is 11.5% over-
all and up to 40% among people 70 years of age
or older.28 By contrast, the prevalence of kidney
failure that requires replacement therapy is 0.2%.
Worldwide, the prevalence of CKD is 9.1% (rep-
resenting approximately 700 million people).29
n engl j med 386;22  nejm.org  June 2, 2022
The New England Journal of Medicine
Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
The annual number of deaths from kidney fail-
ure is 1.2 million, with an additional 1.4 million
deaths due to CVD attributed to CKD; thus, 4.6%
of all deaths are due to CKD (making it the 12th
leading cause of death). These results under-
score the need for detection, evaluation, and
treatment of even early stages of kidney disease
to slow progression and prevent complications.
Risk Factors for CVD
CVD is the most important complication of
CKD. Both a decreased GFR and increased albu-
minuria are independent risk factors for many
manifestations of CVD (Table 1), including coro-
nary heart disease, stroke, heart failure, periph-
eral arterial disease, and need for amputation of
any lower extremity; the strongest associations
are with heart failure and death from CVD.30-32
There is a higher incidence of many CVD events
in CKD than of events of kidney failure requir-
ing replacement therapy. Including the GFR and
the level of albuminuria is as important as in-
cluding traditional risk factors in assessing the
risk of CVD. A lower eGFRcys is more strongly
associated with CVD than is a lower eGFRcr,
which provides additional justification for more
widespread use of cystatin C. Studies show that
the occurrence of CVD events, particularly heart
failure, in patients with CKD hastens the pro-
gression of kidney disease, suggesting a bidirec-
tional relationship between CKD and CVD.33,34
Current guidelines recommend evaluation and
treatment for CKD in patients with CVD and
evaluation and treatment for CVD in patients
with CKD.35
Clinical Trials on Progression of Kidney
Disease
There are few treatments to slow the progres-
sion of kidney disease, in part because it is dif-
ficult to conduct clinical trials of treatment for
CKD. An important obstacle is that in many
kidney diseases, the mean decline in the GFR is
only 2 to 3 ml per minute per 1.73 m2 per year,
requiring a long duration of follow-up to reach
the clinical end point of kidney failure. Use of
predictive biomarkers (markers associated with
an increased risk of the clinical end point) and
surrogate end points (markers that occur earlier
than clinical end points and predict the treat-
ment effect) can help facilitate the conduct of
trials. Both a decreased GFR and increased albu-
2125
 The n e w e ng l a n d j o u r na l
minuria are predictive biomarkers for the pro-
gression of kidney disease and are widely used
as eligibility criteria for enrolling high-risk pa-
tients in clinical trials. The validity of these
measures as surrogate end points depends on
many factors (Table 1 and Table S6).36-38
A decline in the GFR is a step on the path to
kidney failure and is therefore a meaningful end
point for CKD from all causes and for all inter-
ventions. Regulatory bodies now accept as end
points a 30% or 40% decline in the GFR and
mean changes in the GFR (slopes) over a period
of 2 to 3 years in various settings, in addition to
the established surrogate end point of a dou-
bling of the serum creatinine level (equivalent
to a 57% decline in the eGFRcr). An increase in
albuminuria is appealing as a surrogate end
point because it can occur before a decline in the
GFR, often within 6 months after the interven-
tion is begun. However, an increase in albumin-
uria is limited as an end point because it is ap-
propriate only for diseases characterized by
albuminuria, interventions targeting albumin-
uria, and study participants with increased albu-
minuria at baseline and because longer follow-
up may be necessary to determine changes in
the GFR and the safety of the interventions.
There is also interest in using changes in both
albuminuria and the GFR as a combined end
point.39 Trials of recently approved drugs, in-
cluding tolvaptan in polycystic kidney disease,
glucagon-like peptide agonists and finerenone
in CKD with type 2 diabetes, and several inhibi-
tors of sodium–glucose cotransporter 2 in CKD
with or without type 2 diabetes, have shown
successful application of these surrogate end
points for the progression of kidney disease. Trials
involving patients with other chronic diseases in
which CKD is common (e.g., CVD, cancer, infec-
tious diseases, and autoimmune diseases) may
also benefit from the use of these surrogate end
points to evaluate the potential synergistic ef-
fects of therapeutic agents on the progression of
kidney disease.
Risk-Prediction Instruments for CKD
Outcomes
The GFR and albuminuria are used in risk-pre-
diction instruments such as equations to guide
clinical decision making (Table 2). For example,
the Kidney Failure Risk Equation uses the GFR,
2126 n engl j med 386;22  nejm.org  June 2, 2022
The New England Journal of Medicine
Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
albumin-to-creatinine ratio, age, and sex to pre-
dict the 2-year and 5-year risks of kidney failure
with replacement therapy for patients with a
GFR of less than 60 ml per minute per 1.73 m2,
and its routine use is recommended by current
guidelines to guide evaluation and treatment.40
A higher predicted risk can be used to prioritize
referral for a consultation with a nephrologist,
multidisciplinary CKD care, or planning for
kidney-replacement therapy. The GFR and level
of albuminuria can also be incorporated into
existing CVD risk prediction instruments to
modify CVD risk predictions.41 More frequent
use of these and other instruments holds prom-
ise for improving a broad spectrum of clinical
outcomes across the full range of GFR and albu-
minuria values.
Older age predicts a lower risk of kidney fail-
ure that requires replacement therapy in the Kid-
ney Failure Risk Equation and other instruments
because of the competing effect of death, which
emphasizes the difference in prognosis between
older and younger people with CKD. Some peo-
ple have suggested that the GFR threshold for
the definition of CKD should be age-adapted (a
lower GFR threshold for older people than for
younger people) because of the higher preva-
lence of CKD and lower relative risk of kidney
failure among older people.42 In our view, the
absolute and the relative risks associated with all
CKD outcomes, not only kidney failure, should
be considered, and the evaluation and manage-
ment of CKD, not the definition, should be
adapted for age and other risk factors with the
use of appropriate risk-prediction instruments.43
C onclusions
The GFR and level of albuminuria are powerful
measures for detecting and staging AKD and
CKD and for predicting the risk of kidney-­
disease progression and CVD, but they remain
underused. We suggest that greater use of these
measures can lead to further advances in clinical
practice, research, and public health.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
We thank our long-time colleagues Tom Greene, Mark J. Sar-
nak, Josef Coresh, Ron T. Gansevoort, Kunihiro Matsushita, and
Hiddo J. Lambers-Heerspink for their contributions, Yingying
Sang for providing risk predictions, and Ogechi M. Adingwupu
for assistance in the preparation of an earlier version of the
manuscript.
 Uses of GFR and Albuminuria in Kidney Disease

References
1. National Kidney Foundation. K/DOQI
clinical practice guidelines for chronic
kidney disease: evaluation, classification,
and stratification. Am J Kidney Dis 2002;​
39:​Suppl 1:​S1-S266.
2. Levey AS, Eckardt K-U, Dorman NM,
et al. Nomenclature for kidney function
and disease: report of a Kidney Disease:
Improving Global Outcomes (KDIGO)
Consensus Conference. Kidney Int 2020;​
97:​1117-29.
3. Levin A, Agarwal R, Herrington WG,
et al. International consensus definitions
of clinical trial outcomes for kidney fail-
ure: 2020. Kidney Int 2020;​98:​849-59.
4. Kidney Disease: Improving Global
Outcomes (KDIGO) Acute Kidney Injury
Work Group. KDIGO clinical practice
guideline for acute kidney injury. Kidney
Int Suppl 2012;​2:​1-138.
5. Kidney Disease: Improving Global
Outcomes (KDIGO) CKD Work Group.
KDIGO 2012 clinical practice guideline
for the evaluation and management of
chronic kidney disease. Kidney Int Suppl
2013;​3:​1-150.
6. Lameire NH, Levin A, Kellum JA, et al.
Harmonizing acute and chronic kidney
disease definition and classification: re-
port of a Kidney Disease: Improving
Global Outcomes (KDIGO) Consensus
Conference. Kidney Int 2021;​100:​516-26.
7. Ix JH, Shlipak MG. The promise of tu-
bule biomarkers in kidney disease: a review.
Am J Kidney Dis 2021;​78:​719-27.
8. Chawla LS, Bellomo R, Bihorac A, et al.
Acute kidney disease and renal recovery:
consensus report of the Acute Disease
Quality Initiative (ADQI) 16 Workgroup.
Nat Rev Nephrol 2017;​13:​241-57.
9. Benzing T, Salant D. Insights into glo-
merular filtration and albuminuria. N Engl
J Med 2021;​384:​1437-46.
10. Eriksen BO, Palsson R, Ebert N, et al.
GFR in healthy aging: an individual par-
ticipant data meta-analysis of iohexol
clearance in European population-based
cohorts. J Am Soc Nephrol 2020;​31:​1602-
15.
11. Denic A, Lieske JC, Chakkera HA, et al.
The substantial loss of nephrons in healthy
human kidneys with aging. J Am Soc
Nephrol 2017;​28:​313-20.
12. Teo BW, Zhang L, Guh J-Y, et al. Glo-
merular filtration rates in Asians. Adv
Chronic Kidney Dis 2018;​25:​41-8.
13. Delgado C, Baweja M, Crews DC, et al.
A unifying approach for GFR estimation:
recommendations of the NKF-ASN Task
Force on reassessing the inclusion of race
in diagnosing kidney disease. J Am Soc
Nephrol 2021;​32:​2994-3015.
14. Shlipak MG, Tummalapalli SL, Boul-
ware LE, et al. The case for early identifi-
cation and intervention of chronic kid-
ney disease: conclusions from a Kidney
Disease: Improving Global Outcomes
(KDIGO) Controversies Conference. Kid-
ney Int 2021;​99:​34-47.
15. Miller WG, Kaufman HW, Levey AS,
et al. National Kidney Foundation Labora-
tory Engagement Working Group recom-
mendations for implementing the CKD-
EPI 2021 race-free equations for estimated
glomerular filtration rate: practical guid-
ance for clinical laboratories. Clin Chem
2022;​68:​511-20.
16. Levey AS, Coresh J, Tighiouart H,
Greene T, Inker LA. Measured and esti-
mated glomerular filtration rate: current
status and future directions. Nat Rev
Nephrol 2020;​16:​51-64.
17. Sumida K, Nadkarni GN, Grams ME,
et al. Conversion of urine protein-creati-
nine ratio or urine dipstick protein to
urine albumin-creatinine ratio for use in
chronic kidney disease screening and
prognosis: an individual participant-based
meta-analysis. Ann Intern Med 2020;​173:​
426-35.
18. Weaver RG, James MT, Ravani P, et al.
Estimating urine albumin-to-creatinine
ratio from protein-to-creatinine ratio: de-
velopment of equations using same-day
measurements. J Am Soc Nephrol 2020;​
31:​591-601.
19. Susantitaphong P, Cruz DN, Cerda J,
et al. World incidence of AKI: a meta-
analysis. Clin J Am Soc Nephrol 2013;​8:​
1482-93.
20. Levey AS, de Jong PE, Coresh J, et al.
The definition, classification, and prog-
nosis of chronic kidney disease: a KDIGO
Controversies Conference report. Kidney
Int 2011;​80:​17-28.
21. Libório AB, Leite TT, Neves FMO,
Teles F, Bezerra CTM. AKI complications
in critically ill patients: association with
mortality rates and RRT. Clin J Am Soc
Nephrol 2015;​10:​21-8.
22. Inker LA, Grams ME, Levey AS, et al.
Relationship of estimated GFR and albu-
minuria to concurrent laboratory abnor-
malities: an individual participant data
meta-analysis in a global consortium. Am
J Kidney Dis 2019;​73:​206-17.
23. Fox CS, Matsushita K, Woodward M,
et al. Associations of kidney disease mea-
sures with mortality and end-stage renal
disease in individuals with and without
diabetes: a meta-analysis. Lancet 2012;​
380:​1662-73.
24. Mahmoodi BK, Matsushita K, Wood-
ward M, et al. Associations of kidney dis-
ease measures with mortality and end-
stage renal disease in individuals with
and without hypertension: a meta-analy-
sis. Lancet 2012;​380:​1649-61.
25. Chang AR, Grams ME, Ballew SH, et al.
Adiposity and risk of decline in glomeru-
lar filtration rate: meta-analysis of indi-
vidual participant data in a global consor-
tium. BMJ 2019;​364:​k5301.
26. Hallan SI, Matsushita K, Sang Y, et al.
Age and association of kidney measures
with mortality and end-stage renal dis-
ease. JAMA 2012;​308:​2349-60.
27. James MT, Levey AS, Tonelli M, et al.
Incidence and prognosis of acute kidney
diseases and disorders using an integrat-
ed approach to laboratory measurements
in a universal health care system. JAMA
Netw Open 2019;​2(4):​e191795.
28. Levey AS, Stevens LA, Schmid CH, et al.
A new equation to estimate glomerular
filtration rate. Ann Intern Med 2009;​150:​
604-12.
29. GBD Chronic Kidney Disease Collab-
oration. Global, regional, and national bur-
den of chronic kidney disease, 1990–2017:
a systematic analysis for the Global Bur-
den of Disease Study 2017. Lancet 2020;​
395:​709-33.
30. Chronic Kidney Disease Prognosis
Consortium. Association of estimated glo-
merular filtration rate and albuminuria
with all-cause and cardiovascular mortal-
ity in general population cohorts: a collab-
orative meta-analysis. Lancet 2010;​ 375:​
2073-81.
31. Matsushita K, Coresh J, Sang Y, et al.
Estimated glomerular filtration rate and
albuminuria for prediction of cardiovas-
cular outcomes: a collaborative meta-
analysis of individual participant data.
Lancet Diabetes Endocrinol 2015;​3:​514-25.
32. Matsushita K, Ballew SH, Coresh J,
et al. Measures of chronic kidney disease
and risk of incident peripheral artery dis-
ease: a collaborative meta-analysis of in-
dividual participant data. Lancet Diabetes
Endocrinol 2017;​5:​718-28.
33. Sud M, Tangri N, Pintilie M, Levey AS,
Naimark D. Risk of end-stage renal dis-
ease and death after cardiovascular events
in chronic kidney disease. Circulation
2014;​130:​458-65.
34. Sud M, Tangri N, Pintilie M, Levey AS,
Naimark DM. ESRD and death after heart
failure in CKD. J Am Soc Nephrol 2015;​
26:​715-22.
35. Rangaswami J, Bhalla V, Blair JEA,
et al. Cardiorenal syndrome: classification,
pathophysiology, diagnosis, and treatment
strategies: a scientific statement from the
American Heart Association. Circulation
2019;​139(16):​e840-e878.
36. Levey AS, Inker LA, Matsushita K, et al.
GFR decline as an end point for clinical
trials in CKD: a scientific workshop spon-
sored by the National Kidney Foundation
and the US Food and Drug Administra-
tion. Am J Kidney Dis 2014;​64:​821-35.
37. Levey AS, Gansevoort RT, Coresh J,
et al. Change in albuminuria and GFR as
end points for clinical trials in early stages
of CKD: a scientific workshop sponsored
by the National Kidney Foundation in col-
laboration with the US Food and Drug
Administration and European Medicines
Agency. Am J Kidney Dis 2020;​75:​84-104.

n engl j med 386;22  nejm.org  June 2, 2022 2127

The New England Journal of Medicine
Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.
 Uses of GFR and Albuminuria in Kidney Disease

38. Inker LA, Grams M, Gudmandottir H,
et al. Developing treatments for early
stages of common, chronic kidney dis-
eases: clinical trial considerations. Am J
Kidney Dis (in press) (https://www​.­kidney​
.­org/​­news/​­developing​-­t reatments​-­e arly​
-­stages​-­common​-­chronic​-­kidney​-­diseases​
-­clinical​-­t rial​-­considerations).
39. Coresh J, Grams ME, Chen TK. Using
GFR, albuminuria, and their changes in
clinical trials and clinical care. Am J Kid-
ney Dis 2021;​78:​333-4.
40. Tangri N, Grams ME, Levey AS, et al.
Multinational assessment of accuracy of
equations for predicting risk of kidney
failure: a meta-analysis. JAMA 2016;​315:​
164-74.
41. Matsushita K, Jassal SK, Sang Y, et al.
Incorporating kidney disease measures
into cardiovascular risk prediction: devel-
opment and validation in 9 million adults
from 72 datasets. EClinicalMedicine 2020;​
27:​100552.
42. Glassock RJ, Delanaye P, Rule AD.
Should the definition of CKD be changed
to include age-adapted GFR criteria? YES.
Kidney Int 2020;​97:​34-7.
43. Levey AS, Inker LA, Coresh J. “Should
the definition of CKD be changed to in-
clude age-adapted GFR criteria?” Con: the
evaluation and management of CKD, not
the definition, should be age-adapted.
Kidney Int 2020;​97:​37-40.
Copyright © 2022 Massachusetts Medical Society.

images in clinical medicine

The Journal welcomes consideration of new submissions for Images in Clinical
Medicine. Instructions for authors and procedures for submissions can be found
on the Journal’s website at NEJM.org. At the discretion of the editor, images that
are accepted for publication may appear in the print version of the Journal,
the electronic version, or both.

2128 n engl j med 386;22  nejm.org  June 2, 2022

The New England Journal of Medicine

Downloaded from nejm.org by SERGIO PRIETO on June 1, 2022. For personal use only. No other uses without permission.
Copyright © 2022 Massachusetts Medical Society. All rights reserved.