University of Groningen
Novel aspects regarding mechanisms and consequences of albuminuriaScheven, Lieneke
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2014
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Scheven, L. (2014). Novel aspects regarding mechanisms and consequences of albuminuria.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 06-05-2021
Chapter 2Consequences of the use of albumin to creatinine ratio instead of 24-hour urinary albumin excretion for albuminuria staging
Lieneke SchevenPriya VartHiddo J. Lambers HeerspinkPaul E. de JongDick de ZeeuwRon T. Gansevoort
Submitted
20
Abstract
Introduction
New guidelines advocate the use of albumin to creatinine ratio in a urine sample (ACR) instead
of 24h-urinary albumin excretion (UAE) for staging albuminuria. Concern has been expressed
that this may result in misclassification because of among others interindividual differences in
urinary creatinine excretion. We examined the number of participants that are reclassified when
using ACR instead of UAE, the characteristics of reclassified participants and their outcome.
Methods
Included were 6,047 participants of the PREVEND and RENAAL studies of whom data on UAE
and ACR (in a first morning void urine sample) were available at baseline. For categorization
3 albuminuria classes were used as advocated by KDIGO (<30, 30-300, >300 mg/24h and
mg/g).
Results
When using ACR in the first morning void instead of UAE 89% of participants were classified
in corresponding albuminuria categories. 234 (3.9%) participants were classified to a higher
and 426 (7.0%) to a lower category. Participants that were reclassified upward had in general
a worse cardiovascular risk profile when compared to non-reclassified participants, whereas
the reverse was true for downward reclassified participants. In line, Cox regression analyses
showed that upward reclassification when using first morning void ACR instead of UAE was
associated with a tendency for increased risk for cardiovascular morbidity and mortality and
all-cause mortality, whereas downward reclassification was associated with a tendency for
lower risk. The Net Reclassification index, adjusted for age, gender and duration of follow-up,
when using ACR categories instead of UAE categories was 0.068 (p=0.016) for cardiovascular
events and 0.064 (p=0.015) for all-cause mortality.
Conclusion
Our results indicate that although there is reclassification when using first morning void albumin
to creatinine ratio instead of 24h-urinary albumin excretion, reclassification is often correct and
indicative for prognosis.
21
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
Introduction
Elevated albuminuria has been established as a valuable risk marker for renal and cardiovascular
complications (1-5). Albuminuria can be assessed in several ways, of which measurement
of 24-hour urinary albumin excretion (24h UAE) was for long considered the gold standard.
Recent guidelines, however, advocate the use of the albumin to creatinine ratio (ACR) in a spot
urine sample (6). Opponents of the use of ACR to assess albuminuria argue that sometimes a
subject is called having increased albuminuria based on an ACR, whereas a 24h-urine collection
does not support this.
Reasons for discrepancy between albuminuria staging based on 24h UAE versus ACR can
be, first, that an increased ACR may not only be due to an increase in albuminuria, but also
to a decrease in urinary creatinine concentration. Creatinine, as waste product of muscle
catabolism, is dependent on muscle mass and consequently differs by, among others, age
and gender. Second, albuminuria is subject to a circadian rhythm, whereas urinary creatinine
excretion is fairly stable during the day (7). Assessment of the ACR in a first morning urine
sample may therefore reveal another value than the ACR in a 24h-urine collection. Third, 24h
UAE is subject to collection errors and reclassification to another risk category based on spot
urine ACR may therefore also be due to incorrect 24h urine collection.
The aim of this study is to analyze whether misclassification by expressing albuminuria as
ACR instead of 24h UAE plays an important role. Importance is assessed by calculating the
percentage of participants in which it occurs, as well as by studying whether reclassification
reflects clinical characteristics and prognosis in these participants. For these analyses we
used data of two studies in which sera, and 24h-urine and first morning urine samples were
available and allowed calculation of 24h UAE and ACR.
Patients and Methods
This study is a collaboration between the investigators of the Prevention of REnal and Vascular
ENd-stage Disease (PREVEND) study and the Reduction of Endpoints in NIDDM with the
Angiotensin II Antagonist Losartan (RENAAL) study. This collaboration was established
to ensure sufficient participants in all three albuminuria classes of the present CKD staging
system for analyses.
22
Study design and population
The PREVEND study is a prospective cohort study which investigates the natural course of
albuminuria and its relation to renal and cardiovascular disease. Details of the study protocol
have been published elsewhere (9,10). In brief, the participants of the PREVEND study have
been selected in 1997 from subjects of the general population in Groningen, aged 28–75
years. Pregnancy and insulin usage were exclusion criteria. In total, 8,592 subjects participated
in the first screening (1997–98), of which 6,000 participants had a UAC >10 mg/L in the spot
morning urine sample and 2,592 participants a UAC <10 mg/L. This screening consisted
of two outpatient clinic visits, where baseline measurements were performed. Part of this
screening was that participants collected two 24h urines, of which the first sample was used
for the present analyses.
The RENAAL study is a multinational, double-blind randomised placebo-controlled study that
evaluated the renal protective effects of the angiotensin-II blocker losartan in patients with type
2 diabetes and nephropathy. The study design and results have been reported elsewhere (11-
13). In brief, in- and exclusion criteria for the RENAAL study were: type 2 diabetes (assessed
as age over 30 years old at time of diagnosis, no history of ketoacidosis and not using insulin
therapy within 6 months after diagnosis), a serum creatinine between 1.3 and 3.0 mg/dl (1.5 to
3.0 mg/dl for males more than 60 kg), a urinary ACR from a first-morning specimen of at least
300 mg/g, HbA1c<12% and age between 31 and 70 years. All patients collected at baseline a
first-morning void urine sample for albumin and creatinine assessment. In addition, a random
sample of 701 patients collected 24-hour urine samples for quantification of UAE.
Both studies were approved by medical ethics committees and conducted in accordance with
the International Conference of Harmonization Good Clinical Practice Guidelines and adhere to
the ethical principles that have their origin in the Declaration of Helsinki.
For the present analyses we use data of participants in the two studies of whom sera, 24h-urine
and first morning urine samples were available and allowed calculation of 24h UAE and ACR
in a first morning sample. Furthermore, participants with errors in their 24h-urine sampling
were excluded, leaving 6,047 participants for the present study (5,366 of 8,592 PREVEND
participants and 681 of 1,513 RENAAL participants).
Measurements and Definitions
At the baseline visit anthropometrical measurements were performed, blood pressure was
measured and fasting blood and urine samples were taken in which analytes were measured
using routine methodology. Urinary albumin concentration was measured by nephelometry
(PREVEND: BNII, Dade Behring Diagnostic, Marburg, Germany; RENAAL: Beckman Array,
23
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
Beckman, Fullerton, California, USA).
Participants that smoked in the year prior to the baseline screening were regarded smokers.
Cardiovascular disease history was defined by self-report. Hypertension was defined in
accordance with the JNC-7 criteria as systolic blood pressure (SBP) of ≥140 mmHg or diastolic
blood pressure (DBP) ≥90 mmHg or use of antihypertensive medication. Diabetes was defined
in accordance with the ADA criteria as a fasting glucose level of >7.0 mmol/L or non-fasting
glucose level of >11.1 mmol/L or use of anti-diabetic medication. Body mass index (BMI) was
calculated as the ratio between weight and the square of height (weight/height2). Errors in
24h-urine collections were defined as the upper and lower 2.5% of the difference between the
estimated creatinine excretion rate and the actually measured creatinine excretion rate. The
estimated creatinine excretion rate was calculated by: eCER = 879.89 + 12.51 * weight (kg) -
6.19 * age + (34.51 if black) - (379.42 if female) (14).
Albuminuria measures
24h UAE (mg/24h) is given as urinary albumin concentration times volume of one 24h-urine
collection. ACR (mg/g) was calculated by dividing urinary albumin concentration (mg/L) by
urinary creatinine concentration (g/L). Cut-off values indicating normoalbuminuria, and
moderately and severely increased albuminuria were defined in accordance with the prevailing
KDIGO CKD guideline (6): for 24h UAE <30, 30-300 and >300 mg/24h, and for ACR <30, 30-
300 and >300 mg/g, respectively.
Outcome
The outcomes assessed were cardiovascular morbidity and mortality and all-cause mortality
during follow-up. In PREVEND information on date and cause of death was obtained by
record linkage with the Dutch Central Bureau of Statistics. Information on hospitalization for
cardiovascular morbidity was obtained from PRISMANT, the Dutch national registry of hospital
discharge diagnoses. In RENAAL information on cardiovascular morbidity and mortality was
collected prospectively. For this study, incident cardiovascular morbidity was defined according
the Major Adverse Cardiovascular Events (MACE) criteria (11)
Statistical analysis
Baseline characteristics were calculated for the overall population and for participants stratified
according to the three UAE categories. Continuous data are reported as mean ± standard
deviation (SD). In case of non-parametric data distribution, medians with interquartile ranges
(IQR) are presented. Differences in baseline characteristics between the three UAE subgroups
were calculated with a Chi-square test for categorical data and for continuous data with a one-
24
way ANOVA (in case of non-parametric data a Kruskal–Wallis ANOVA by ranks test).
To assess reclassification, we created 3x3 cross-tabulations using the aforementioned clinically
relevant cut-off values for 24h UAE and ACR. In these tables the proportion of reclassified
participants was calculated. The McNemar test, a nonparametric test for comparing two
related samples, was used to test the significance of the ratios of up and down classification
between 24h UAE versus ACR. Differences in characteristics between non-reclassified and
reclassified pariticpants in table 2 were calculated with a Chi-square test for categorical data,
and for continuous data with Student’s t-test (in case of normally distributed data) or Mann
Whitney test (in case of non-parametric data).
In addition, we assessed whether risk of cardiovascular morbidity and mortality, and all-
cause mortality differed between participants reclassified and those not reclassified. For risk
assessment Cox regression analyses were used, first crude, second adjusted for age and
gender, and subsequently also for baseline UAE and strata for participation in the PREVEND or
RENAAL study. Follow-up time was defined as the period to the first outcome or loss to follow-
up. Individuals who were free of these outcomes by December 31, 2000 in the RENAAL study
and by January 1, 2009 in the PREVEND study, were subject to administrative censoring.
To assess reclassification we calculated net reclassification improvement (NRI) for cardiovascular
morbidity and mortality and all-cause mortality (15). Net reclassification improvement (NRI)
is a statistic that allows calculation of the effect of reclassification of individuals from one
disease category to the other. It is a difference of two ratios; clinically correct reclassification
minus clinically incorrect classification. The range of this difference is from −1 to +1 with a
negative number reflecting incorrect reclassification and a positive number indicating correct
reclassification. Given the difference in follow-up duration in PREVEND and RENAAL, we
calculated NRI adjusted for age, gender as well as for duration of follow-up. NRI was also
calculated per subgroup of age (above and below the median age in event cases), gender, BMI
(<30 and ≥30 kg/m2), and hypertension and diabetes mellitus status.
All calculations were performed with SPSS (version 18.0) and STATA (version 12.0) software. A
p-value of less than 0.05 was considered to indicate statistical significance.
25
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
Results
Baseline characteristics
Table 1 shows the baseline characteristics of the 6,047 participants. Mean age of the total
cohort was 50.8 ±12.9 years and half of the participants were female. When albuminuria was
assessed as 24h UAE (mg/24h) or as ACR (mg/g), the prevalence of moderately increased
albuminuria was 11.9% and 11.2%, respectively, and the prevalence of severely increased
albuminuria 11.7% and 10.5%, respectively. Table 1 also shows that when participants had
higher UAE their cardiovascular risk profile was higher, i.e. these participants were older,
more often male, more obese, and had a higher prevalence of hypertension and diabetes.
Differences in characteristics between in- and excluded PREVEND participants, and between
in- and excluded RENAAL participants, are shown in Supplementary Table 1. These differences
were in general numerically very small, although because of the large number of participants
they sometimes reached statistical significance.
Reclassification when using ACR instead of 24h UAE
When using ACR instead of 24h UAE 89% of participants were classified in corresponding
albuminuria categories (Table 2). 234 (3.9%) participants were classified to a higher and 426
(7.0%) to a lower category (Table 2). Participants who were reclassified upward to a higher
albuminuria category when using ACR instead of 24h UAE (for example when comparing 24h
UAE <30mg/24h and reclassification to a higher ACR category) were older in comparison to
participants that were not reclassified (52.6 ±13.0 vs. 48.1 ±12.5 yrs, p<0.001), and more often
of female gender (74.4 vs. 53.0%, p<0.001), diabetic (4.0 vs 1.7, p=0.004) and hypertensive
(35.5% vs 28.4%, p=0.017). Furthermore, they had a higher body mass index (26.4 vs. 25.7
kg/m2, p=0.014), and tended to have more often a cardiovascular disease history (6.2% vs.
4.3%, p=0.068). In contrast, the participants reclassified downward to a lower albuminuria
stage (for example when comparing 24h UAE >300mg/24h and reclassification to a lower
ACR category) were less often of female gender (26.8 vs. 37.7%, p=0.041), diabetic (66.7
vs. 92.6%, p<0.001) and hypertensive (94.2 vs. 98.2%, p=0.013). Furthermore, they had less
often a cardiovascular disease history (42.9 vs. 75.0% p<0.001) (Table 2).
Clinical consequences of reclassification
The 6,047 participants had a median follow-up time of 10.5 [6.2-10.8] years (a total of 51,509
person years), during which 732 events of cardiovascular morbidity and mortality and 1,082
events of all-cause mortality occurred. Participants that were reclassified upward had higher
event rates and participants that were reclassified downward had lower event rates when
compared to non-reclassified participants. Cox regression analyses showed that upward
26
Tab
le 1
. Bas
elin
e ch
arac
teris
tics
of s
tudy
pop
ulat
ion
over
all a
nd a
ccor
ding
to 2
4h a
lbum
inur
ia s
tage
s
Uri
nary
Alb
umin
Exc
reti
on
(mg
/24h
)
Ove
rall
<30
30-3
00>
300
(N=
6,04
7)(n
=4,
619)
(n=
720)
(n=
708)
p-v
alue
Age
(yrs
)50
.8 ±
12.9
48.3
±12
.657
.6 ±
12.1
59.8
±8.
1<
0.00
1
Mal
e ge
nder
(%)
49.9
45.7
63.5
63.8
0.00
7
Bod
y m
ass
inde
x (k
g/m
2 )26
.4 ±
4.6
25.7
±3.
927
.9 ±
4.7
29.9
±6.
2<
0.00
1
Sm
okin
g (%
)35
.537
.439
.319
.5<
0.00
1
His
tory
of c
ardi
ovas
cula
r di
seas
e (%
)8.
04.
415
.723
.7<
0.00
1
Ser
um g
luco
se (m
g/dL
)84
.7 (7
7.5-
95.5
)82
.9 (7
7.5-
90.1
)90
.1 (8
1.1-
106.
3)15
1.0
(111
.0-2
08.0
)<
0.00
1
Use
of g
luco
se lo
wer
ing
med
icat
ion
(%)
7.3
0.9
8.5
47.6
<0.
001
Dia
bete
s m
ellit
us (%
)13
.91.
817
.988
.4<
0.00
1
Sys
tolic
blo
od p
ress
ure
(mm
Hg)
131.
6 ±
21.8
126.
4 ±
18.7
145.
1 ±
23.9
151.
9 ±
20.8
<0.
001
Use
of a
ntih
yper
tens
ive
med
icat
ion
(%)
21.9
12.1
31.3
76.3
<0.
001
Hyp
erte
nsio
n (%
)40
.928
.768
.592
.5<
0.00
1
Ser
um c
hole
ster
ol (m
g/dL
)21
9.0
±45
.321
2.4
(185
.3-2
43.2
)22
3.9
(196
.9-2
51.9
)22
2.0
(194
.0-2
60.0
)0.
019
Ser
um c
reat
inin
e (m
g/dL
)1.
05 ±
0.4
0.93
±0.
161.
09 ±
0.39
1.78
±0.
67<
0.00
1
Ser
um a
lbum
in (g
/dL)
4.5
±0.
44.
6 ±
0.3
4.5
±0.
33.
8 ±
0.5
0.00
2
24-h
our
urin
e
- U
rine
albu
min
con
cent
ratio
n (m
g/L)
7.4
(4.2
-19.
9)5.
6 (3
.4-9
.1)
42.5
(28.
3-73
.9)
707.
5 (3
56.0
-147
0.0)
<0.
001
- U
rinar
y al
bum
in e
xcre
tion
(mg/
24h)
9.8
(6.3
-27.
1)7.
9 (5
.8-1
1.7)
63.4
(41.
7-11
0.6)
1378
.0 (6
79.5
-285
4.0)
<0.
001
- U
rine
crea
tinin
e co
ncen
trat
ion
(g/L
)0.
98 ±
0.4
0.98
±0.
51.
00 ±
0.4
0.84
±0.
40.
001
- A
lbum
in-c
reat
inin
e ra
tio (m
g/g)
7.6
(4.9
-20.
5)6.
1 (4
.5-9
.0)
46.3
(29.
2-89
.3)
1068
.9 (5
23.7
-242
1.8)
<0.
001
Firs
t-m
orni
ng v
oid
- U
rine
albu
min
con
cent
ratio
n (m
g/L)
12.6
(7.4
-25.
9)11
.1 (5
.6-1
4.7)
34.7
(19.
5-61
.1)
665.
0 (3
06.5
-137
2.5)
<0.
001
- U
rine
crea
tinin
e co
ncen
trat
ion
(g/L
)1.
37 ±
0.7
1.51
±0.
71.
16 ±
0.6
0.71
±0.
4<
0.00
1
- A
lbum
in-c
reat
inin
e ra
tio (m
g/g)
8.3
(5.0
-22.
2)6.
5 (4
.4-1
0.4)
33.5
(18.
5-68
.6)
996.
5 (4
48.4
-230
2.3)
<0.
001
27
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
Table 2. Characteristics of study participants by urinary albumin categories using the 24h urinary albumin
excretion
(mg/24h) and the albumin to creatinine ratio (mg/g)
ACR (mg/g)
<30 30-300 >300
UAE (mg/24h) (n=4,734) (n=678) (n=635)
<30 (n=4,619)
No. 4416 198 5
Age (years) 48.1 ± 12.5 52.6 ± 13.0 NR
Male (%) 46.6 25.8 NR
Body mass index (kg/m2) 25.7 ± 3.9 26.4 ± 4.7 NR
Smoking (%) 37.9 31.3 NR
CVD history (%) 4.3 6.2 NR
Diabetes mellitus (%) 1.7 4.0 NR
Hypertension (%) 28.4 35.5 NR
30-300 (n=720)
No. 317 372 31
Age (years) 54.6 ± 12.8 59.7 ± 11.2 63.6 ± 7.4
Male (%) 67.8 60.8 51.6
Body mass index (kg/m2) 27.5 ± 4.7 28.1 ± 4.6 29.3 ± 6.5
Smoking (%) 42.6 38.7 12.9
CVD history (%) 11.7 19.5 69.2
Diabetes mellitus (%) 6.4 22.9 77.4
Hypertension (%) 59.3 75.3 100
>300 (n=708)
No. 1 108 599
Age (years) NR 60.3 ± 8.3 59.7 ± 8.1
Male (%) NR 73.2 62.3
Body mass index (kg/m2) NR 30.5 ± 4.8 29.8 ± 6.4
Smoking (%) NR 23.4 18.9
CVD history (%) NR 42.9 75.0
Diabetes mellitus (%) NR 66.7 92.6
Hypertension (%) NR 94.2 98.2
UAE= urinary albumin excretion, ACR=albumin creatinine concentration, CVD=cardiovascular disease,
NR=not reliable because of low numbers
28
reclassification when using ACR instead of 24h UAE was associated with a significantly
increased risk of cardiovascular morbidity and mortality, and all-cause mortality (Table 3). For
instance, for cardiovascular morbidity and mortality, upward reclassification from 24h UAE
moderately increased albuminuria (30-300 mg/24h) to an ACR severely increased albuminuria
(>300 mg/g) was associated with a crude Hazard Ratio of 2.83 (95% CI 1.46- 5.48) and
when adjusted for age, gender, baseline 24h UAE and cohort of 2.86 (95% CI 1.32- 6.16).
Downward reclassification was associated with a tendency to lower risk in crude analyses,
but not after adjustment for age and gender (Table 3). Similar results were obtained for all-
cause mortality. When using the NRI to analyze whether participants were reclassified upward
correctly to a higher cardiovascular risk category based on their ACR (i.e. that more of these
participants had a cardiovascular event during follow-up), it showed that the NRI adjusted
for age, gender and duration of follow-up was significantly positive i.e. 0.068 (0.013- 0.123),
p=0.016. NRI for all-cause mortality was 0.064 (0.013- 0.114), p=0.015. Figures 1 and 2
show NRIs for cardiovascular events and for all-cause mortality, respectively, across various
subgroups. NRIs for risk categorization using ACR instead of 24h UAE were positive or similar
in all subgroups, with lower values obtained in subgroups characterized by lower muscle
mass. For cardiovascular events the NRI was significantly lower for females when compared
to males (Figure 1), and for all-cause mortality again for females when compared to males, as
well as for age 60 or more when compared to less than 60 years (Figure 2).
Discussion
For practical reasons the ACR has been advocated as the standard method to assess
albuminuria for chronic kidney disease (CKD) staging instead of 24h UAE, which was
traditionally regarded to be the gold standard to assess albuminuria. This advice has led to
concern whether it may lead to misclassification and especially overdiagnosis of CKD, because
of the creatinine component in the ACR, and that upward reclassification by using ACR may
not reflect patient prognosis. We investigated these issues in the present study.
Various studies have previously compared timed urinary albumin excretion rates (in 24h or
overnight samples) and ACR, and investigated which ACR values correspond with the
traditionally used cut-off values for timed urinary albumin excretion rates to define CKD (16,17).
In general these studies are difficult to interpret because of the methodological shortcoming
that ACR was determined in the same urine sample as the timed urinary albumin excretion
rate (18). It is generally acknowledged that urinary albumin excretion follows a circadian rhythm
(19,20). This rhythm is dependent on among others posture, exercise and dietary factors (such
29
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
Tab
le 3
. C
linic
al o
utco
mes
acc
ordi
ng to
recl
assi
ficat
ion
stat
us b
y al
bum
in to
cre
atin
ine
ratio
(AC
R, m
g/g)
com
pare
d w
ith 2
4h u
rinar
y al
bum
in e
xcre
tion
(UA
E, m
g/24
h)
Car
dio
vasc
ular
mo
rbid
ity
and
mo
rtal
ity
All-
caus
e m
ort
alit
y
By
AC
R (m
g/g
) rec
lass
ifica
tio
nB
y A
CR
(mg
/g) r
ecla
ssifi
cati
on
Do
wnw
ard
No
neU
pw
ard
Do
wnw
ard
No
neU
pw
ard
24h
UA
E (m
g/2
4h)
(n=
89)
(n=
608)
(n=
35)
(n=
127)
(n=
905)
(n=
50)
<30
Num
ber
of e
vent
sN
A34
125
NA
528
34
Cru
de in
cide
nce
rate
NA
8.2
13.8
NA
12.7
18.8
Cru
de h
azar
d ra
tioN
AR
efer
ence
1.71
(1.1
4 –
2.57
)N
AR
efer
ence
1.49
(1.0
6 –
2.11
)
Adj
uste
d ha
zard
rat
ioa
NA
Ref
eren
ce1.
61 (1
.07
– 2.
43)
NA
Ref
eren
ce1.
37 (0
.97
– 1.
95)
Adj
uste
d ha
zard
rat
iob
NA
Ref
eren
ce1.
44 (0
.96
– 2.
18)
NA
Ref
eren
ce1.
24 (0
.87
– 1.
76)
30-3
00
Num
ber
of e
vent
s58
9110
9213
816
Cru
de in
cide
nce
rate
21.6
31.9
98.9
34.2
48.3
157.
8
Cru
de h
azar
d ra
tio0.
68 (0
.49
– 0.
95)
Ref
eren
ce2.
83 (1
.46
– 5.
48)
0.71
(0.5
5 –
0.93
)R
efer
ence
3.14
(1.8
6 –
5.31
)
Adj
uste
d ha
zard
rat
ioa
0.85
(0.6
1 –
1.19
)R
efer
ence
2.78
(1.4
2 –
5.43
)0.
87 (0
.67
– 1.
14)
Ref
eren
ce3.
03 (1
.78
– 5.
15)
Adj
uste
d ha
zard
rat
iob
0.84
(0.5
8 –
1.21
)R
efer
ence
2.86
(1.3
2 –
6.16
)0.
89 (0
.67
– 1.
20)
Ref
eren
ce2.
85 (1
.55
– 5.
25)
>30
0
Num
ber
of e
vent
s31
176
NA
3523
9N
A
Cru
de in
cide
nce
rate
55.9
90.9
NA
63.2
123.
4N
A
Cru
de h
azar
d ra
tio0.
72 (0
.49
– 1.
07)
Ref
eren
ceN
A0.
58 (0
.40
– 0.
83)
Ref
eren
ceN
A
Adj
uste
d ha
zard
rat
ioa
0.70
(0.4
7 –
1.04
)R
efer
ence
NA
0.56
(0.3
9 –
0.80
)R
efer
ence
NA
Adj
uste
d ha
zard
rat
iob
0.85
(0.5
6 –
1.28
)R
efer
ence
NA
0.70
(0.4
8 –
1.03
)R
efer
ence
NA
a =
adj
uste
d fo
r ag
e an
d ge
nder
; b =
as
a+ a
dditi
onal
adj
ustm
ent f
or b
asel
ine
UA
E a
nd c
ohor
t
30
Figure 1. Net Reclassification Improvements (NRIs) for Cardiovascular Morbidity and Mortality for various
subgroups (*p<0.05 for difference in NRIs).
Figure 2. Net Reclassification Improvements (NRIs) for All-Cause Mortality for various subgroups (*p<0.05
for difference in NRIs).
31
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
as protein and fluid intake), with lower values during nighttime than during daytime. It is therefore
to be expected that the cut-off values of ACR indicating microalbuminuria will be different when
these variables are measured in a spot morning sample instead of a 24-hour urine collection.
Consequently, ACR should be measured in a specifically collected spot or first morning void
urine sample and compared with a timed urinary albumin excretion sample. Only few studies
followed such a design, but these studies did not investigate what the consequences are for
CKD staging or risk classification when spot ACR is used instead of timed urinary albumin
excretion rates (21,22). Our study adds therefore novelty to existing literature.
Our findings indicate that using ACR instead of 24h UAE does not lead to more participants
being labeled as having CKD. Only a limited percentage of participants is reclassified (10.9%),
especially to a lower albuminuria stage (7.0%). The participants that were reclassified to a higher
albuminuria stage (3.9%) had in general a higher cardiovascular and renal risk profile, and an
increased risk for cardiovascular events as well as all-cause mortality when compared to non-
reclassified participants. The NRI when using ACR instead of 24h UAE for risk categorization
was also found to be significantly positive for cardiovascular events as well as for all-cause
mortality. In stratified analyses NRIs tended to be lower in subgroups characterized by lower
muscle mass, but were positive or similar in all subgroups. One may argue therefore that
participants have been correctly reclassified using ACR instead of 24h UAE. This assumption
is supported by our finding that reclassification seems for a part the result of errors in 24h urine
collection and therefore in errors in 24h UAE assessment. In participants reclassified upward
from the 24h UAE albuminuria stage 30-300 mg/24h to ACR stage >300 mg/g expected 24h
creatinine excretion was 13.8 (8.6- 15.9) mmol, whereas actually measured creatinine excretion
was only 7.8 (5.1- 10.9) mmol (p<0.05), suggesting that 24h urine collection was incomplete.
Our data thus do not provide evidence for overdiagnosis of CKD, nor for inadequate risk
stratification in reclassified subjects when using ACR instead of 24h UAE. As such these data
support the prevailing KDIGO guidelines that advise to use an ACR in a first morning void or
spot urine sample to assess albuminuria for CKD staging.
This study has limitations that should be acknowledged. First, this analysis comprises
participants of two separate studies, PREVEND and RENAAL, that differ in design. PREVEND
is a prospective, observational, general population based cohort study, whereas RENAAL is a
double-blind, randomized, placebo-controlled trial investigating the effect of the angiotensin-2
blocker losartan versus placebo in patients with type 2 diabetes and nephropathy. We
accounted therefore in our Cox regression analyses for differences in design and participant
characteristics using strata for cohort. Second, despite the combination of the two studies, still
some cells in the cross-tabulation tables included a relatively low number of participants, which
32
consequently leads to limited power when analyzing differences between non-recategorized
and recategorized participants. Third, we used for our analyses first morning void urine samples,
whereas in clinical practice often spot urine samples are used. It has been suggested that first
morning void urine samples may be more reliable than a spot urine sample to diagnose and
monitor increased albuminuria (21). Whether our results will be similar when using spot urine
samples instead of first morning voids needs therefore additional study.
Strengths of this study are the availability of a large number of participants with higher levels
of albuminuria, and that in participants both 24 UAE as well as ACR in a first morning void
were assessed. Only very few other epidemiological studies, if any, collected these data
simultaneously. Furthermore, albuminuria was measured in fresh urine samples like in clinical
practice, and not in samples that have been stored frozen. It has previously been shown that
frozen storage of urine samples leads to a decrease in average urinary albumin concentration,
with more variability, which negatively affects the prognostic value of albuminuria (23).
Furthermore, in study participants data on incident cardiovascular events and mortality were
collected prospectively. Lastly, by design we excluded subjects in whom it was likely that
24h urine collection was inadequate. Even in this scenario using ACR was as strong, or even
stronger, for risk categorization when compared to using 24h UAE, which makes our findings
robust.
In conclusion, our results indicate that reclassification when using ACR instead of 24h UAE
is limited, that reclassification is more often downward than upward, leading to a smaller
number of subjects being labelled as having CKD, and that upward reclassified subjects
have a worse prognosis with respect to incident cardiovascular events and mortality and
downward reclassified subjects have a lower risk. These data thus do not provide evidence for
overdiagnosis of CKD or to misclassification with respect to prognosis in reclassified subjects
when using ACR instead of 24h UAE. Consequently our findings support the advice of the
prevailing KDIGO guidelines to use ACR in a first morning void or spot urine sample to stage
CKD.
Disclosures
The PREVEND study was financially supported by several grants from the Dutch Kidney
Foundation. In this study measurement of urinary albumin concentration was sponsored by
Dade Behring Diagnostica, Marburg, Germany and Roche Diagnostics, Mannheim, Germany.
The RENAAL study was sponsored by Merck & Co, Inc, USA.
33
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
References
1. Chronic Kidney Disease Prognosis Consortium, Matsushita K, van der Velde M, Astor BC, Woodward
M, Levey AS, et al. Association of estimated glomerular filtration rate and albuminuria with all-cause
and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet
2010 Jun 12;375(9731):2073-2081.
2. Gansevoort RT, Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, et al. Lower
estimated GFR and higher albuminuria are associated with adverse kidney outcomes in both general
and high-risk populations. A collaborative meta-analysis of general and high-risk population cohorts.
Kidney Int 2011 Feb 2.
3. Astor BC, Matsushita K, Gansevoort RT, van der Velde M, Woodward M, Levey AS, et al. Lower
estimated glomerular filtration rate and higher albuminuria are associated with mortality and end-
stage renal disease. A collaborative meta-analysis of kidney disease population cohorts. Kidney Int
2011 Feb 2.
4. van der Velde M, Matsushita K, Coresh J, Astor BC, Woodward M, Levey A, et al. Lower estimated
glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular
mortality. A collaborative meta-analysis of high-risk population cohorts. Kidney Int 2011 Feb 9.
5. Levey AS, de Jong PE, Coresh J, El Nahas M, Astor BC, Matsushita K, et al. The definition,
classification, and prognosis of chronic kidney disease: a KDIGO Controversies Conference report.
Kidney Int 2011 Jul;80(1):17-28.
6. http://www.kdigo.org/clinical_practice_guidelines/pdf/CKD/KDIGO_2012_CKD_GL.pdf.KDIGO
Guidelines 2013.
7. Koopman MG, Krediet RT, Koomen GC, Strackee J, Arisz L. Circadian rhythm of proteinuria:
consequences of the use of urinary protein:creatinine ratios. Nephrol Dial Transplant 1989;4(1):9-14.
8. Ellam TJ, El Nahas M. Proteinuria thresholds are irrational: a call for proteinuria indexing. Nephron Clin
Pract 2011;118(3):c217-24.
9. Mahmoodi BK, Gansevoort RT, Veeger NJ, Matthews AG, Navis G, Hillege HL, et al. Microalbuminuria
and risk of venous thromboembolism. JAMA 2009 May 6;301(17):1790-1797.
10. Lambers Heerspink HJ, Brantsma AH, de Zeeuw D, Bakker SJ, de Jong PE, Gansevoort RT, et al.
Albuminuria assessed from first-morning-void urine samples versus 24-hour urine collections as a
predictor of cardiovascular morbidity and mortality. Am J Epidemiol 2008 Oct 15;168(8):897-905.
11. Brenner BM, Cooper ME, de Zeeuw D, Grunfeld JP, Keane WF, Kurokawa K, et al. The losartan
renal protection study--rationale, study design and baseline characteristics of RENAAL (Reduction of
Endpoints in NIDDM with the Angiotensin II Antagonist Losartan). J Renin Angiotensin Aldosterone
Syst 2000 Dec;1(4):328-335.
12. Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, et al. Effects of losartan on
renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med
2001 Sep 20;345(12):861-869.
13. Bakris GL, Weir MR, Shanifar S, Zhang Z, Douglas J, van Dijk DJ, et al. Effects of blood pressure
level on progression of diabetic nephropathy: results from the RENAAL study. Arch Intern Med 2003
Jul 14;163(13):1555-1565.
34
14. Ix JH, Wassel CL, Stevens LA, Beck GJ, Froissart M, Navis G, et al. Equations to estimate creatinine
excretion rate: the CKD epidemiology collaboration. Clin J Am Soc Nephrol 2011 Jan;6(1):184-191.
15. Pencina MJ, D’Agostino RB, Steyerberg EW: Extensions of net reclassification improvement
calculations to measure usefulness of new biomarkers. Statist Med 2011, 30: 11-21.
16. Dyer AR, Greenland P, Elliott P, Daviglus ML, Claeys G, Kesteloot H, et al. Evaluation of measures of
urinary albumin excretion in epidemiologic studies. Am J Epidemiol 2004 Dec 1;160(11):1122-1131.
17. Lambers Heerspink HJ, Gansevoort RT, Brenner BM, Cooper ME, Parving HH, Shahinfar S, et al.
Comparison of different measures of urinary protein excretion for prediction of renal events. J Am Soc
Nephrol 2010 Aug;21(8):1355-1360.
18. Gansevoort RT, Brinkman J, Bakker SJL, de Jong PE, de Zeeuw D: Evaluation of Measures of Urinary
Albumin Excretion. American Journal of Epidemiology 2006, 164: 725-727.
19. Hansen HP1, Tauber-Lassen E, Jensen BR, Parving HH: Effect of dietary protein restriction on
prognosis in patients with diabetic nephropathy. Kidney Int. 2002 Jul;62(1):220-8.
20. van Acker BA1, Stroomer MK, Gosselink MA, Koomen GC, Koopman MG, Arisz L: Urinary protein
excretion in normal individuals: diurnal changes, influence of orthostasis and relationship to the renin-
angiotensin system. Contrib Nephrol. 1993;101:143-50.
21. Witte EC, Lambers Heerspink HJ, de Zeeuw D, Bakker SJL, de Jong PE, Gansevoort R: First
Morning Voids Are More Reliable Than Spot Urine Samples to Assess Microalbuminuria. Journal of
the American Society of Nephrology 2009, 20: 436-443.
22. Chavan VU, Durgawale PP, Sayyed AK, Sontakke AV, Attar NR, Patel SB, Patil SR, Nilakhe SD.: A
Comparative Study of Clinical Utility of Spot Urine Samples with 24-h Urine Albumin Excretion for
Screening of Microalbuminuria in Type 2 Diabetic Patients. Indian J Clin Biochem. 2011 Jul;26(3):283-
9.
23. Brinkman JW, de Zeeuw D, Gansevoort RT, Duker JJ, Kema IP, de Jong PE, Bakker SJ: Prolonged
Frozen Storage of Urine Reduces the Value of Albuminuria for Mortality Prediction. Clin Chem. 2007
Jan;53(1):153-4.
35
Con
sequ
ence
s of
the
use
of a
lbum
in to
cre
atin
ine
ratio
inst
ead
of 2
4-ho
ur u
rinar
y
albu
min
exc
retio
n fo
r al
bum
inur
ia s
tagi
ngC
hapt
er 2
Sup
ple
men
tary
Tab
le 1
. Bas
elin
e ch
arac
teris
tics
of in
clud
ed a
nd e
xclu
ded
PR
EV
EN
D a
nd R
EN
AA
L st
udy
part
icip
ants
PR
EV
EN
DR
EN
AA
L
Incl
uded
(n=
5,36
6)E
xclu
ded
(n=
3,22
6)p
-val
ueIn
clud
ed(n
=68
1)E
xclu
ded
(n=
832)
p-v
alue
Age
(yea
rs)
49.5
± 1
2.9
48.7
± 1
2.2
0.00
159
.9 ±
7.4
60.0
± 7
.50.
342
Mal
e (%
)48
53<
0.00
162
640.
182
Bod
y M
ass
Inde
x (k
g/m
2 )25
.9 ±
4.1
26.3
± 4
.40.
001
30.0
± 6
.329
.4 ±
6.2
0.04
1
Sm
okin
g (%
)38
380.
5717
190.
24
Ser
um g
luco
se (m
g/dL
)85
(77
– 92
)86
(79
– 95
)0.
0716
2 (1
23-2
14)
158
(120
-213
)0.
54
Use
of g
luco
se lo
wer
ing
med
icat
ion
(%)
1.5
1.5
0.47
100
100
1.00
Dia
bete
s m
ellit
us (%
)3
40.
014
100
100
1.00
Sys
tolic
blo
od p
ress
ure
(mm
Hg)
129
± 2
112
9 ±
20
0.68
151
± 2
015
4 ±
19
0.00
3
Use
of a
ntih
yper
tens
ive
med
icat
ion
(%)
1513
0.03
610
010
01.
00
Hyp
erte
nsio
n (%
)34
330.
1310
010
01.
00
Ser
um c
hole
ster
ol (m
g/dL
)21
8 ±
44
218
± 4
20.
6222
5 ±
53
230
± 5
70.
035
Ser
um c
reat
inin
e (m
g/dL
)0.
95 ±
0.2
50.
95 ±
0.1
70.
511.
83 ±
0.5
1.89
± 0
.50.
18
Urin
ary
albu
min
exc
retio
n (m
g/24
hrs)
8.9
(6.2
-15.
9)10
.6 (6
.5-2
0.2)
0.46
1362
(596
-290
2)13
58 (5
92-2
900)
0.56