REVIEW

Screening for Microalbuminuria: Which

Measurement?
S.M. Marshall
Depxtnient of' Medicine, University of' Newcastle upon Tyne, UK
It now seems worth while to identify Type 1 diabetic patients with microalbuminuria, as
improved blood glucose control and reduction of arterial blood pressure will slow if not
prevent the progression to persistent proteinuria. Measurement of albumin excretion rate
(AER) in a timed urine sample remains the gold standard for the definition of
microalbuminuria, but i s not a practical screening procedure. Thus attempts have been
made to relate the albumin concentration or a1bumin:creatinine ratio in random or first
morning urine samples to AER. There i s a weak correlation of albumin concentration
( r = 0.32 to 0.68) and a1bumin:creatinine ratio ( r = 0.43 to 0.54) in a random urine
sample with AER, and low sensitivity and specificity of a variety of different albumin
concentrations and a1bumin:creatinine ratios to predict microalbuminuria. The correlation
of albumin concentration ( r = 0.86 to 0.90) and a1bumin:creatinine ratio ( r = 0.74 to
0.91) in an early morning urine sample with AER i s stronger. Measurement of
a1bumin:creatinine ratio in an early morning urine sample appears to be the most reliable
method of screening for microalbuminuria, with sensitivity of 88 to 100 % and specificity
81 to 100 % depending on the cut-off ratio chosen and the definition of microalbuminuria
used. I f the a1bumin:creatinine ratio in an early morning urine sample i s ~ 3 . mg
5 mmol-',
the patient can be classed as normoalbuminuric and re-screened annually. I f the ratio i s
3 1 0.0 mg mmol-', confirmation of microalbuminuria should be sought in a timed urine
collection and appropriate therapy begun. If the ratio i s 3.6 to 9.9 mg mmol-', there is
a high risk of more rapid progression to definite microalbuminuria and the patient should
be re-screened more frequently.

KEY WORDS Albumin excretion rate Microalbuminuria Nephropathy

Introduction the initial papers showing the predictive power of

Implementation of a screening programme for a disease
state requires that the outcome of the condition can be
improved by early detection and that there is a screening
test available which i s cheap, simple, reliable, and
accurate. There is now good evidence that the presence
of microalbuminuria in Type 1 diabetic patients predicts
the later development of end-stage renal d i s e a s e , ' ~ ~
and in Type 2 diabetic patients, early death from
cardiovascular causes.5," Improved blood glucose con-
tro17,t3
and reduction of arterial blood pressure'' will at least
slow if not prevent progression from microalbuminuria to
persistent proteinuria in Type 1 patients. The effect
microalbuminuria used timed urine collections" and
defined microalbuminuria on the basis of a calculated
albumin excretion rate (AER). The urine sample collected
ranged from 24 h,3,4 through ~ v e r n i g h t ' , to
~ 2-3 h
collections2 (Table 1). Albumin was assayed by a variety
of laboratory methods, the normal reference ranges
quoted varied widely, and the AER above which patients
subsequently developed end-stage renal disease was also
different. It i s particularly important to observe that in
some cases the lower limit of the albumin excretion
rate which was associated with an increased risk of
progression was well above the upper limit of the normal
reference range, implying that some patients have
' , * s 4

of these and other therapeutic manoeuvres, including
modification of other cardiovascular risk factors, in
patients with Type 2 diabetes i s not known. Thus it now
seems worth while to identify at least those Type 1
diabetic patients with microalbuminuria. However, the
best way of doing this is not clear.
Early Studies
There are almost as many definitions of microalbuminuria
as there are papers on the subject. All but one of
Correspondence to: Dr S.M. Marshall, Department of Medicine, The
Medical School, Framlington Place, Newcastle upon Tyne, NE2 4HH,
UK
albumin excretion rates which are abnormal but have
not been demonstrated to be at risk of progression to
persistent proteinuria. In an attempt to overcome these
difficulties and to help standardize research, a consensus
has been reached"' defining microalbuminuria as an AER
of 20-200 p,g min-' in an overnight collection, or
30-300 mg 24-h-'. However, it must be remembered
that these definitions are arbitrary and that any division
into categories is rather artificial because albumin
excretion i s a continuous rather than a discrete variable.
No studies comparing the predictive powers of the
overnight and 24-h albumin excretion rates have been
performed, so both remain gold standards against which
other measures must be validated.

706 0 742-3 0 7 11911080706-06 $05.00

0 1991 by John Wiley & Sons, Ltd DIABETIC MEDICINE, 1991 ; 8: 706-71 1
 Table 1. Summary of the studies showing the predictive power of microalbuminuria in
Type 1 diabetic patients

Viberti et a/.' Mogensen and Parving et a/.' Mathiesen et a/.4

Christensen'

Number of patients 63 43 23 71
Length of follow-up (years) 14 >7 6 6
Type of urine sample overnight short day-time 24-h 24-h
Number of samples 21 23 1 1
Albumin assay RIA RIA RID RID
Reference range AER ~ 1 k2g min-' ~ 7 . 5pg min-' S40 mg 24-h-' C 2 0 pg min-'
Discriminant AER 30 pg min-' 15 pg min-' 40 mg 24-h-' 70 pg min-'
Number above discriminant 718 12114 618 717
level developing proteinuria
Number below discriminant 2155 0129 211 5 3164
level developing proteinuria

RIA, radioinimunoassay; RID, radial irnrnunodiffusion

Analytical Methods the albumin concentration in a random urine sample as

A comprehensive review of analytical methods has
already been published.' I Quantitative immunoassays
specific for albumin are preferred, provided they achieve
the desired sensitivity and specificity. The choice of
method i s likely to depend on locally available expertise
and equipment. If no laboratory assay is available, then
one of the commercially produced, semi-quantitative
side-room tests may be used. Latex bead immunoagglutin-
ation" and the imrnunochemical reagent strip test' ' are
both specific for albumin.
Practical Methods
Whilst measurement of the albumin excretion rate in a
timed urine sample remains essential for research work,
it is obviously not suitable for large-scale screening,
being cumbersome for patients and laboratory staff, and
open to inaccuracies in completeness of urine collection
and sample timing. Attempts have therefore been made
to relate either the albumin concentration or albumin:
creatinine ratio in random or first morning urine samples
to the AER.
Random Urinary Albumin Concentration
Gatling and colleagues reported a correlation coefficient
of 0.45 between albumin concentration measured in a
random urine sample and AER in a timed overnight urine
sample.'' An albumin concentration > 25 mg I-] had
a sensitivity of 56 % and specificity 81% of predicting
an AER > 30 k g m i n - ' . In a small study of children,15
only four of whom had a 24-h urinary albumin excretion
rate > 30 kg min-', the albumin concentration in spot
urine samples collected at various times of the day
correlated with the 24-h AER with a correlation coefficient
of 0.32-0.68. Nathan and colleagues reported that using
SCREENIN(; FOR MlCROALBUMlNURlA
an index of 24-h albumin excretion had sensitivity
ranging from 55 to 83 % depending on the cut-off points
used.I6 In a study of 160 Type 1 patients, an albumin
concentration > 15 mg I - ] in a random urine sample
had a 96 Yo sensitivity and an 80 % specificity of
identifying an AER in a timed overnight collection of
> 15 k g min-I, and an albumin concentration > 26 mg
I-' a sensitivity of 100 % and a specificity of 85 % of
predicting AER > 30 k g min-'.I7
Thus the albumin concentration in a random urine
sample appears to correlate poorly with AER in a
timed urine sample. Despite the differences in samples,
laboratory methods and cut-off points, it seems clear
from the above studies that simple measurement of
albumin concentration in a random urine sample is not
a good screening test to identify diabetic patients with
microalbuminuria. The low specificity in particular
implies a large number of false positive results, generating
needless anxiety and considerable work in repeat testing
or performing timed studies.
Response to Exercise
It i s well recognized that the increase in albumin excretion
in response to exercise is exaggerated in patients with
diabetes.l8 Some" but not a 1 P studies have suggested
that the change in albumin excretion in response to
exercise may be a more sensitive way of identifying
patients with microalburninuria. However, the prognostic
significance of exercise-induced proteinuria in diabetes
is not known and, indeed, in the non-diabetic population,
postural and exercise-induced proteinuria are regarded
as benign conditions. However, the lack of correlation
of the albumin concentration in a random urine sample
with the overnight or 24-h AER has been ascribed to
exercise-induced increases in albumin excretion. Thus,
to negate the effects of exercise and possibly provide a
707
more consistent measurement, it has been suggested that
an early morning urine sample should be used.
Early Morning Urine Albumin
Concentration
The correlation of the albumin concentration in a first
morning urine sample to the albumin excretion rate
in a 24-h” or overnight c ~ l l e c t i o n ’appears
~ higher
(0.86-0.90) than that of the albumin concentration in a
random urine sample. Cowell and colleagues reported a
sensitivity of 100 % and specificity 57 % of albumin
concentration > 20 mg I-’ to predict 24-h AER > 20
mg 24-hkI.” In a much larger study, an albumin
concentration > 20 mg I-’ had a sensitivity of 86 % and
specificity of 97 % of predicting overnight AER > 30 p g
m i n - I . ‘-I Using similar cut-off values, Marshall and
Alberti found a sensitivity of 91 % and specificity 74 % . 2 1
In the largest study reported, the correlation coefficient
of the albumin concentration measured in a first morning
urine sample with the timed overnight AER was 0.90.2L
Using an albumin concentration > 17 mg I-’ to predict
an AER > 30 p g min-’, the sensitivity was 97 % and
specificity 91 %. In contrast to these studies, Wiegmann
and colleagues reported much lower sensitivity (71 %)
and specificity (23 %) of an albumin concentration > 30
mg Ikl to predict 24-h AER > 20 p g mink’.23
Apart from this last study, the overall impression i s
that the correlation of albumin concentration in a first
morning urine sample to AER is stronger than that of
albumin concentration in a random urine sample. An
albumin concentration greater than 17-20 mg Ikl in an
early morning urine sample is a more sensitive and
specific predictor of an elevated AER than an albumin
concentration measured in a random urine sample.
However, the specificity and sensitivity of the test are
still significantly less than 100 %, implying large numbers
of false positive and negative results. Thus further
refinements to the screening test have been suggested,
particularly correction for urine flow rate by measuring
creatinine in the same sample and calculating the
a1bumin:creatinine ratio.
Albumin:Creatinine Ratio in Random
Urine Samples
Using random urine samples, both Gatling et a/.14 and
Cowell et a / . I 5 found only a weak correlation of
a1bumin:creatinine ratio with AER (I = 0.43 to 0.54).
However in a small study, an a1bumin:creatinine ratio
> 2.3 mg mmol-’ had a sensitivity of 94 % and a
specificity of 96 % of identifying an albumin excretion
rate > 15 k g mink’ and a ratio > 3.4 mg mmol-’ had
100 % sensitivity and specificity of identifying AER > 30
pg min-I.’” In several larger studies the results were
less good, an a1bumin:creatinine ratio > 3.4 mg mmolk’
having an 82 % sensitivity and an 81 % specificity of
708
identifying a 24-h AER > 20 p g Gatling and
colleagues produced similar figures of 80 % and 81 Yo
for a ratio > 3.0 mg mmol-’ predicting AER > 30 p g
mink’ in an overnight c ~ l l e c t i o n . ’Perhaps
~ the best
figures were produced by Watts and colleagues who
reported a sensitivity of 100 Yo and a specificity of 80 Yo
of an a1bumin:creatinine ratio 2 2 . 9 mg mmolk’ in
identifying an AER > 15 pg min-’, and a ratio > 8.0
having sensitivity of 100 YO and specificity of 96 % of
predicting an AER > 30 p g mink’.” Thus although
the correction of albumin concentration for creatinine
excretion in a random urine sample does increase the
sensitivity and specificity of the measurement as an index
of albumin excretion rate, the overall results remain
disappointing.
A/bumin:Creatinine Ratio in Early Morning
Urine Sample
Correction of the albumin concentration in a first morning
urine sample also seems to improve the screening test.
Cowell and colleagues reported a correlation coefficient
of 0.74 for a1bumin:creatinine ratio and 24-h albumin
excretion rate,15 and Gatling and co-workers a coefficient
of 0.91 for overnight excretion rate.14 In the hands of
Gatling and colleagues, using an a1bumin:creatinine ratio
> 3.5 mg mmol-’ to predict an overnight AER > 30 p g
mink’ gave a sensitivity of 100 % and specificity of
95 % in one study,l4 and 88 and 99 %, respectively, in
another Lowering the ratio to 2.0 mg mmolk’
increased the sensitivity to 96 % and the specificity to
100 In a small study, Cohen and colleagues reported
that an a1bumin:creatinine ratio cut-off of 2.5 mg mmol-’
produced clear separation of normal and microalbumin-
uric subjects.*’ Another study gave a sensitivity of 98 70
and a specificity of 69 % of an a1bumin:creatinine ratio
> 3.5 mg mmolk’ of identifying an AER > 30 bg
min-’.*’ If the ratio was increased to 4.5 mg mmol-’,
the sensitivity was 96 % and specificity 80 %.
Conclusions
Drawing definitive conclusions from these stcrdies is
difficult if not impossible, given that almost all have
used different cut-off points for albumin concentration,
a1bumin:creatinine ratio, and albumin excretion rates.
The use of an albumin concentration in a random urine
sample as a screening test for microalbuminuria i s not
to be recommended because of unacceptably low
sensitivity and specificity. Also, the intra-subject day-to-
day variability in albumin excretion in a daytime urine
sample is greater than in an overnight sample, presumably
because of the effect of variable exercise.Lb This variation
would make it difficult to interpret changes with time in
the screening test. It thus seems advisable to use an early
morning urine sample, which can be collected under
more easily standardized conditions, and thus may more
5 M MARSHALL
 Dm
readily reflect pathological rather than physiological
changes in albumin excretion. However, the price to
pay for this may be lower compliance, random urine
samples being much easier to collect during a clinic
visit.
The choice therefore lies between measurement of the
albumin concentration or the a1bumin:creatinine ratio in
the early morning urine sample. In an attempt to compare
like with like, all of the studies defining microalbuminuria
as an albumin excretion rate > 30 p g min-' in a timed
overnight urine collection are detailed in Table 2. This
cut-off is the one used most frequently in the reported
studies and has been shown to be predictive of future
nephropathy in Type 1 patients' and early death in Type
2 patients.' Once again, the lack of consistency in the
reported sensitivities and specificities makes the choice
REVIEW
anyone with an AER > 30 I J . ~min-', so these patients
simply need to be re-screened, perhaps annually.
The difficulty lies in those patients with a1bumin:creati-
nine ratios 3.6-9.9 mg mmolk', of whom approximately
half will have an AER > 30 p.g min-'. As probably 10 %
of the diabetic population lies within this range,'4,12 the
logistics of collecting a timed urine specimen from them
all are huge. It i s likely that those patients with the
z : . :
10 . 'j. . :...
. -.. .. .'....-. .
* :*-
2 - -.t.
;.-5 6 -
. .j

... .
CI
. :
between simple measurement of albumin concentration 2 4-
or calculation of the a1bumin:creatinine ratio difficult. 0) 3-
. . . . . .
... . '
. ...*-**.
* .
: .
.. .:
. . ,. . . . . , , , , .

.-c
Using only the albumin concentration has the advantage .c 2
4- - .
t .*.. . :
of being cheaper, but there i s little information on various m
cut-off points. More studies have been published using
the a1bumin:creatinine ratio, but as Table 2 illustrates,
it is impossible to choose between the various albumin:
creatinine cut-off values. s 01
a
Perhaps one way out ot the difficulty is to adopt a +I 1 1 1 1 1
two-tier interpretation of the a1bumin:creatinine ratio as 0 10 30 100 200 500
has previously been ~ u g g e s t e d . ' ~Ratios ~ 1 0 . 0mg Albumin excretion ratelpg min-'I
mmol- are clearly found almost exclusively in those
patients with an AER > 30 p g min-l (Figure l ) . 1 4 Figure 1 . Relationship of albumin excretion rate in a timed
overnight urine sample to a1bumin:creatinine ratio in 129
These patients should have confirmatory timed albumin diabetic patients; : albumin excretion rate of 30 p,g min-',
excretion rates measured and appropriate therapy started. __ a1bumin:creatinine ratio of 10.0 mg mmol-I,. . albumin:.
Ratios less than 3.5 mg rnmol-' are highly unlikely in creatinine ratio of 3.5 mg rnmol-'

Table 2. Studies using albumin excretion rate of 30 kg min-I in a timed overnight urine
sample as definition of microalbuminuria

Reference Sample Measurement Cut-off Sensitivity Specificity

used point (%I (%)
~ ~~

Gatling et a1.l4 random concentration >25 56 81

Watts et a/.'' random concentration >26 100 85
Gatling et a/.24 random ratio >3.0 80 81
Watts et random ratio X.0 100 96
Gatling et a/,14 EMU concentration >20 86 97
Marshall and EMU concentration 320 91 74
AlbertiL'
Hutchison et a/.'' EMU concentration >17 97 91
Gatling et a/.'4 EMU concentration >20 82 96
Gatling et a\." EMU ratio >2.0 96 100
Cohen et EMU ratio >2.5 100 100
Hutchison et a/.z2 EMU ratio >3.0 97 94
Gatling et EMU ratio >3.5 100 95
Gatling et a/.L" EMU ratio >3.5 88 99
Marshall and EMU ratio >3.5 98 69
'
Al berti'
Marshall and EMU ratio >4.5 96 80
Alberti' I

EMU, early morning u r i i e sample; concentration, albumin concentration irng I- I); ratio,
a1bumin:creatinine ratio (rng rnmol-'1.

SCREENING FOR MICROALBUMINURIA 709

 REVlEW
initially higher 'normal' albumin concentrations will
progress most quickly to m i c r ~ a l b u r n i n u r i aThus
. ~ ~ further
rises in the a1bumin:creatinine ratio would suggest
progression towards niicroalbuniinuria and determine the
need for assessment by timed urine collection. It would
thus appear sensible to repeat the measurement of the
a1bumin:creatinine ratio more frequently, perhaps every
3-6 months, in these patients. It i s likely that this strategy
will at least initially fail to identify some patients with
a1bumin:creatinine ratios of 3.6-9.9 mg mmolk' but
AER > 30 k g niin-I. This may not be vital, as several
papers have suggested that it is those patients with AEK
in the high microalbuminuric range, AEK > 70 or 100
k g min-I, who will progress to Albustix positive
proteinuria4," and who will benefit most from improve-
ment of blood glucose control.8 Most patients with
a1bumin:creatinine ratios of 3.6-9.9 mg mmol-' had
AER < 100 k g min-.' (Figure l ) . 1 4 A 5-year follow-
up of young insulin-treated patients with an initial
albun1in:creatinine ratio >2.5 mg mmolk' has shown a
strong correlation ( r = 0.91) between the initial ratio
and the mean ratio thereafter.'8
Recommendations
A suggested screening strategy i s outlined in Figure 2.
Because of the large day-to-day variation in albumin
excretion,2" more than one estimation should be recorded
before the patient is designated as having microalbumin-
uria. Testing should be performed when the patient is in
stable metabolic control, as acute elevations in blood
glucose cause transient increases in albumin excretion.29
In addition, high concentrations of acetoacetate cause
interference with the Jaff6 method of measurement of
creatinine. w Patients with an initial a1bumin:creatinine
ratio ~ 3 . mg
5 mmol-' can be considered to have normal
albumin excretion and be re-screened annually. Those
with an a1bumin:creatinine ratio 210.0 mg mmolkl
should have a timed albumin excretion rate measured
Initial A1burnin:Creatinine Ratio
+ ++
Microalbuminuria
Rescreen Rescreen Confirmatory
annually at each visit' timed collection
Appropriate treatment
Figure 2. Suggested screening plan for microalbuminuria
71 0
Dm
to confirm the presence of microalbuminuria before
appropriate treatment i s begun. Those in the 'grey' area
with a ratio 3.6-9.9 mg mmol-', should have the ratio
measured more regularly to detect progression towards
definite m icroalbum in uria.
It i s obvious from this review that any screening plan
i s of necessity somewhat arbitrary and that further
longitudinal studies are required to reveal the natural
history of microalbuminuria in diabetes. All research
work should be based on the gold standard of multiple
timed overnight or 24-h urine collections.
References
1. Viberti GC, Hill RD, Jarrett RJ, Argyropoulos A, Mahmud
U, Keen H. Microalbuminuria as a predictor of clinical
nephropathy in insulin-dependent diabetes mellitus. Lan-
cet 1982; i : 1430-1432.
2. Mogensen CE, Christensen CK. Predicting diabetic nephro-
pathy in insulin-dependent patients. N Engl 1 Med 1984;
311: 89-93.
3. Parving HH, O x e n b d l B, Svendsen PAa, Sandahl Christi-
ansen J, Andersen AR. Early detection of patients at risk
of developing diabetic nephropathy. A longitudinal study
of urinary albumin excretion. Acta Endocrinol (Copenhi
1982; 100: 550-555.
4. Mathiesen ER, Oxenb0ll B, Johansen K, Svendsen PAa,
Deckert T. Incipient nephropathy in Type 1 (insulin-
dependent) diabetes. Diabetologia 1984; 26: 406-41 0.
5. Jarrett RJ, Viberti GC, Argyropoulos A, Hill RD, Mahmud
U, Murrells TJ. Microalbuminuria predicts mortality in
non-insulin-dependent diabetes. Diabetic Med 1984; 1 :
17-20.
6. Mogensen CE. Microalbuminuria predicts clinical pro-
teinuria and early mortality in maturity-onset diabetes. N
Engl] M e d 1984; 310: 356-360.
7. Feldt-Rasmussen B, Mathiesen ER, Deckert T. Effect of
two years of strict metabolic control on progression of
incipient nephropathy in insulin-dependent diabetes.
Lancet 1986; ii: 1300-1 304.
8. Feldt-Rasmussen B, Mathiesen ER, Jensen T, Lauritzen T,
Deckert T. Effect of improved metabolic control on loss
of kidney function in Type 1 (insulin-dependent) diabetic
patients: an update of the Steno studies. Diabetologia
1991 ; 34: 164-1 70.
9. Marre M, Chatellier G, Leblanc H, Guyene TT, Menard
J, Passa P. Prevention of diabetic nephropathy with
enalapril in normotensive diabetics with microalbuminu-
ria. Br M e d I 1988; 297: 1092-1 095.
10. Mogensen CE, Chachati A, Christensen CK, Deckert T,
Hommel E, Kastrup J, et a / . Microalbuminuria: an early
marker of renal involvement in diabetes. Uremia Invest
1985; 9: 85-92.
11. Rowe DJF, Dawnay A, Watts GF. Microalbuminuria in
diabetes mellitus: review and recommendations for the
measurement of albumin in urine. Ann Clin Biochem
1990; 27: 297-312.
12. Viberti GC, Vergani D. Detection of potentially reversible
diabetic albuminuria. A three-drop agglutination test for
urinary albumin at low concentration. Diabetes 1982;
31: 973-975.
13. Marshall SM, Shearing PA. Assessment of Micral-Test
strips as a screening tool for microalbuminuria (Abstract).
Diabetic M e d 1991; 8 (suppl 1): in press.
14. Gatling W, Knight C, Hill RD. Screening for early diabetic
S.M. MARSHALL
 DTT7
nephropathy: Which sample to detect microalbuminuria?
Diabetic Me d 1985; 2: 451-455.
1 5 . Cowell CT, Rogers S, Silink M. First morning urinary
albumin concentration is a good predictor of 24-hour
urinary albumin excretion in children with Type 1 (insulin-
dependent) diabetes. Diabetologia 1986; 29: 97-99.
16. Nathan DM, Rosenbauni C, Protasowicki VD. Single-
void urine samples can be used to estimate quantitative
niicroalbuminuria. Diabeteh Care 1987; 10: 414-41 8.
17. Watts GF, Shaw KM, Polak A. The use of random urine
samples to screen for microalbuminuria in the diabetic
clinic. Practical Diabetes 1986; 3: 86-88.
18. Mogensen CE, Vittinghus E. Urinary albumin excretion
during exercise in juvenile diabetes. S a n d 1 Lab lnvest
1975; 35: 295-300.
19. Watts CF, Williams I, Morris RW, Mandalia S, Shaw
KM, Polak A. An acceptable exercise test to study
microalbuminuria in Type 1 diabete3. Diabetic Med 1989;
6: 787-792.
20. Feldt-Rasniussen B, Baker L, Deckert T. Exercise as a
provocative test in early renal disease in Type 1 (insulin-
dependent) diabetes: albuminuric, systemic and renal
iaeniodynaniic responses. Diabetologia 1985; 28:
389-396.
21. Marshall SM, Alberti KGMM. Screening for early diabetic
nephropathy. Ann Clin Biochem 1986; 23: 195-1 97.
22. Hutchison AS, O’Reilly DStJ, MacCuish AC. Albumin
excretion rate, albumin concentration, and albuminkreati-
nine ratio compared for screening diabetics for slight
albuniinuria. Clin Chern 1988; 34: 201 9-2021.
SCREENING F O R MICROALBUMINURIA
REVIEW
23. Wiegmann TB, Chonko AM, Barnard MJ,et al. Compari-
son of albumin excretion rate obtained with different
times of collection. Diabetes Care 1990; 13: 864-871.
24. Gatling W, Knight C, Mullee MA, Hill RD. Microalbumin-
uria in diabetes: a population study o f the prevalence
and an assessment of three screening tests. Diabetic Med
1988; 5 : 343-347.
25. Cohen DL, Close CF, Viberti GC. The variability of
overnight urinary albumin excretion in insulin-dependent
diabetic and normal subjects. Diabetic M e d 1987; 4:
437-440.
26. Chachati A, von Frenckell R, Foidart-Willems J, Godon
JP, Lefebvre PI. Variability of albumin excretion in insulin-
dependent diabetes. Diabetic M e d 1987; 4: 441-445.
27. Messent 1. Progression to microalbuminuria in normoalbu-
niinuric Type 1 (insulin-dependent) diabetic patients
(Abstract). Diabetologia 1990; 33(suppl): A30.
28. Dunn P), Jury DR. Random urine a1bumin:creatinine
ratio measurements as a screening test for diabetic
microalbuminuria-a five year follow up. NZ M e d l 1990;
103: 562-564.
29. Parving H H , Noer I, Deckert T, Evrin PE, Nielsen SL,
Lyngsae J, et a / . The effect of metabolic regulation on
microvascular permeability to small and large molecules
in short-term juvenile diabetics. Dfabetdogia 1976; 12:
161-166.
30. Watts GF, Pillay D. Effect of ketones and glucose on
the estimation of urinary creatinine: implications for
niicroalbuminuria screening. Diabetic Med 1990; 7:
263-2 65.
71 1