CORE CURRICULUM IN NEPHROLOGY

Renal Function Testing

Mitchell H. Rosner, MD, and W. Kline Bolton, MD

䡲 Age: age-related decline in GFR, 0.75 to

ASSESSMENT OF GLOMERULAR
1.0 mL/min/1.73 m2 (0.01 to 0.02 mL/s/
FILTRATION RATE
1.73 m2) per year
䡲 Pregnancy: GFR elevated as much as
Why Test Renal Function? 50% in first trimester and onward; re-
● Patients with kidney disease have few signs turns toward normal by 4 to 8 weeks
and symptoms early in disease course; postpartum
laboratory evaluation may be only way of 䡲 Protein intake: GFR higher in patients on
detecting disease high-protein diet
䡲 Diurnal variation: values tend to be about
● Tests should detect abnormalities early
10% higher in afternoon than at night
enough to allow corrective therapy
䡲 Antihypertensive therapy: secondary to
● Important for measuring renal disease pro-
lowering of blood pressure; variable ef-
gression and efficacy of therapies
fect not directly predictable
● Help predict when renal replacement therapy 䡲 States associated with hyperfiltration: dia-
may be necessary betes, obesity, acromegaly
● Aid in appropriate dosing of medications ● Clearance (C): the rate at which an indica-
● Tests that best detect abnormalities of renal tor substance is removed from plasma per
function measure glomerular filtration rate unit concentration; specifies a volume from
(GFR) which all of a substance is removed per unit
time
Measurement of GFR ● For a substance Z cleared by renal elimina-
● GFR cannot be measured directly tion:
● Remarkably constant in single individual
under constant conditions Cz ⫽ UzxV ⁄ Pz
● Large variation between individuals with
large spread of normal values Where Uz is urinary concentration of z, Pz
● Causes of interpatient variability include: is plasma concentration of Z, and V is urine
flow rate
䡲 Body size: GFR conventionally factored
● If substance z is freely filtered and only
by 1.73 m2
excreted by GFR, then: GFR ⫽ UzxV/Pz
䡲 Sex: GFR approximately 8% higher in
● Thus, plasma concentration of indicator is
males
inversely related to GFR and GFR can be
䡲 Race
assessed from plasma concentration
● Requires an ideal filtration marker (Table 1)
From the Department of Internal Medicine, Division of
to ensure that elimination of substance is
Nephrology, University of Virginia Health System, Charlottes- completely dependent on GFR
ville, VA.
Received July 6, 2005; accepted in revised form August Use of Exogenous Markers to Measure GFR
16, 2005.
Originally published online as doi:10.1053/j.ajkd.2005.08.038 ● Techniques:
on December 5, 2005. 䡲 Urine collection with time-averaged
Address reprint requests to Mitchell H. Rosner, MD, plasma concentration of marker sub-
Division of Nephrology, Box 800133, University of Virginia stance, or
Health System, Charlottesville, VA 22908. E-mail: mhr9r@ 䡲 Plasma clearance determined by either a
virginia.edu
© 2005 by the National Kidney Foundation, Inc. continuous infusion or bolus method
0272-6386/05/4701-0022$30.00/0 ● Marker substances:
doi:10.1053/j.ajkd.2005.08.038 䡲 Inulin (molecular weight [MW], 5,000 d):

174 American Journal of Kidney Diseases, Vol 47, No 1 (January), 2006: pp 174-183
CORE CURRICULUM IN NEPHROLOGY 175

Table 1. Characteristics of an Ideal Marker for 䡲 Females: 21.9 – 0.115 ⫻ age

GFR Measurement ● Serum levels can be affected by numerous
Constant rate of production (or for exogenous marker
factors (Table 2)
can be delivered intravenously at a constant rate) ● Not an ideal marker since it also is excreted
Freely filterable at the glomerulus (minimal protein by tubular secretion:
binding) 䡲 Proportion of total creatinine clearance
No tubular reabsorption (Ccr) due to tubular secretion increases as
No tubular secretion
No extrarenal elimination or metabolism
GFR decreases and Ccr leads to GFR
Availability of an accurate and reliable assay overestimation by approximately 10 mL/
For exogenous marker: safe, convenient, readily min/1.73 m2 (0.17 mL/s/1.73 m2)
available, inexpensive, and does not influence GFR 䡲 In some subjects (eg, those with sickle
(physiologically inert) cell anemia), this GFR overestimation
can be much greater
EFructose polymer, freely filtered, nei- ● Because of reciprocal relationship between
ther reabsorbed nor secreted by tu- GFR and serum creatinine (SCr), a large
bules, physiologically inert change in GFR is initially required to raise
E Determinations are labor intensive, SCr levels from normal; however, once SCr
limiting its widespread use is elevated, small changes in GFR will raise
䡲 Radionuclide-labeled markers: it exponentially more
E Clearance is determined as amount of ● 1/Scr better reflects magnitude of GFR
indicator injected divided by inte- decline
grated area of plasma concentration ● Laboratory determination of creatinine:
curve over time (as determined by 䡲 Jaffé reaction: creatinine reacts with
formulas) picrate under alkaline conditions to form
125
E Most commonly used: I-iothalamate a chromogen:
51
and Cr-ethylenediaminetetra-acetic E Positive interference from glucose,
acid (EDTA) (plasma levels) or 99m-Tc- ascorbic acid, uric acid, acetoacetate,
mercaptoacetyltriglycine (MAG3) pyruvate, ketoacids; results in creati-
(gamma counter) nine value 20% higher than true value
䡲 Radiocontrast markers: E Negative interference; high serum biliru-
E Iothalamate sodium, iohexol (safest), bin levels may cause spuriously low
diatrizoate meglumine; measure io- SCr values
dine levels 䡲 Enzymatic kinetic alkaline picrate method;
less interference from noncreatinine chro-
Use of Endogenous Markers to Determine GFR mogens and much more accurate
● Creatinine (MW, 113 d) 䡲 Enzymatic assays (amidohydrolase
● Metabolism: method) have similar precision to the
䡲 Generated in muscle by nonenzymatic kinetic method
conversion of creatine and phosphocreati-
Table 2. Factors Affecting SCr
nine
䡲 Generation is proportional to muscle Increase SCr Decrease SCr
mass and is relatively constant
䡲 Important role of liver in formation of Ketotic states, hyperglycemia Dietary protein restriction
(Jaffé)* Muscle wasting,
creatinine through methylation of guani- Cephalosporins (Jaffé)* malnutrition
dine aminoacetic acid Flucytosine (enzymatic Bilirubin (Jaffé)*
● Levels vary according to diurnal and men- method)* Renal disease
strual variations, race, and diet (and method Cimetidine, trimethoprim Advanced age
of meat preparation) (block secretion) Female sex
Vigorous exercise Advanced liver disease
● Excretion rate (urinary creatinine excretion Ingesting cooked meats
⫻ V) mg/kg/d:
䡲 Males: 28.2 – 0.172 ⫻ age *These factors affect the measurement process directly.
176 ROSNER AND BOLTON

䡲 All methods lose precision in lower E Overestimates Ccr in vegetarians and

range of assay; thus, SCr is insensitive in patients who are malnourished, on
determining small changes of GFR from low-protein diet, obese, or edematous
normal E In patients with renal disease, pre-
䡲 Autoanalyzers often are calibrated to differ- dicted Ccr overestimates GFR at low
ent creatinine standards with substantial values
variation (up to 0.4 mg/dL [35 ␮mol/L]) 䡲 Modification of Diet in Renal Disease
between creatinine values from different (MDRD) equation predicts GFR more
laboratories accurately than measured Ccr and is
● Creatinine clearance: preferred method: GFR ⫽ 170 ⫻ [SCr
䡲 Calculated from 24-hour urine sample (mg/dL)]⫺0.999 ⫻ [age]⫺0.176 ⫻ [0.762
and single SCr (assumes steady state) if patient is female] ⫻ [1.18 if patient is
䡲 Inaccuracies result from incomplete urine black]:
collections E Not validated across diverse ethnic
䡲 Not valid for patients not in steady state populations, in patients age ⬎60 years
䡲 Overestimates GFR due to tubular secre- or ⬍18 years, with diabetes
tion of creatinine (especially important at E May underestimate GFR in stage 1
lower GFRs) chronic kidney disease and overesti-
䡲 Cimetidine blocks tubular secretion of mate GFR in stages 4 and 5
creatinine and thus increases accuracy of E Failure to calibrate creatinine assay to
Ccr; optimal dosing is not clear and laboratory that developed the estimat-
blockade of tubular secretion often in- ing equation can introduce systematic
complete, thus limiting this technique error in estimated GFR, particularly at
● SCr as a marker of GFR: a high GFR
䡲 Elevated level usually indicates reduced 䡲 Among children, the Schwartz and Cou-
GFR nahan-Barratt formulae provide clini-
䡲 Normal level does not exclude possibil- cally useful estimates of GFR
ity of reduced GFR: 䡲 Note: Use of SCr to estimate GFR relies
E Factors may keep creatinine genera- on steady state; in certain circumstances,
tion rate low (Table 2) clearance methods may be more reliable:
E SCr may thus remain within normal E Extremes of age and body size
ranges in these patients despite a sig- E Severe malnutrition or obesity
nificant GFR reduction E Diseases of skeletal muscle
E Especially important in patients with E Paraplegia or quadriplegia
chronic kidney disease where patients E Vegetarian diet
restrict protein intake and may blunt E Rapidly changing SCr
rise in SCr despite falls in GFR ● Urea (MW, 60 d):
E Liver disease where SCr may be low 䡲 One of first indicators used to measure
despite poor GFR GFR
● Formulae for estimating GFR using SCr: 䡲 Shares few features of ideal marker and
䡲 Equations include factors related to creat- is poor measure of GFR
inine generation and excretion (age, sex, 䡲 Freely filtered but reabsorbed in proximal
race, body size) and distal nephron (urea clearance is less
䡲 Rely on patients being in steady state than GFR); urea reabsorption is substantial
䡲 Major limitation is due to variation in in states of decreased renal perfusion
measurement of SCr with kinetic Jaffé 䡲 Urea production is variable and largely
reaction and autoanalyzers and from in- dependent on protein intake
trinsic variability of SCr 䡲 Many variables affect urea level (Table 3)
䡲 Cockcroft-Gault equation predicts Ccr as 䡲 State of diuresis affects urea clearance
Ccr ⫽ (140 ⫺ age) ⫻ (weight)/(72 ⫻ more than Ccr and is useful in differential
SCr) (multiply by 0.85 if female): diagnosis of acute renal failure (ARF)
CORE CURRICULUM IN NEPHROLOGY 177

Table 3. Factors Affecting Serum Urea Nitrogen filtration rate from serum creatinine: A new prediction
equation. Ann Intern Med 130:461-470, 1999
Increase Serum Urea Decrease Serum Urea 5. Grubb AO: Cystatin C—Properties and use as diagnos-
tic marker. Adv Clin Chem 35:63-99, 2000
Dehydration Volume expansion 6. Knight EL, Verhave JC, Spiegelman D, et al: Factors
Reduced renal perfusion Pregnancy influencing serum cystatin C levels other than renal function
(heart failure) SIADH and the impact on renal function measurement. Kidney Int
Increased dietary protein Restriction of dietary 65:1416-1421, 2004
Catabolic states: protein 7. Coresh J, Astor BC, McQuillan G, et al: Calibration
Fever Liver disease and random variation of the serum creatinine assay as
Trauma Advanced renal disease critical elements of using equations to estimate glomerular
GI bleeding filtration rate. Am J Kidney Dis 39:920-929, 2002
Tetracyclines 8. Cockcroft DW, Gault MH: Prediction of creatinine
Corticosteroids clearance from serum creatinine. Nephron 16:31-34, 1976
9. Luke RG: Urea and the BUN. N Engl J Med
Abbreviations: GI, gastrointestinal; SIADH, syndrome of 305:1213-1215, 1981
inappropriate secretion of antidiuretic hormone. 10. Herrington D, Drusano G, Smalls U, et al: False
elevation in serum creatinine levels. JAMA 252:2962, 1984
where blood urea nitrogen–creatinine ratio (letter)
is increased when causes are prerenal 11. Ibrahim H, Mondress M, Tello A, et al: An alternative
● Cystatin C (MW, 13,000 d): formula to the Cockcroft-Gault and the Modification of Diet
in Renal Diseases formulas in predicting GFR in individuals
䡲 Cysteine proteinase inhibitor produced with type 1 diabetes. J Am Soc Nephrol 16:1051-1060, 2005
by all nucleated cells at constant rate 12. Froissart M, Rossert J, Jacquot C, et al: Predictive
䡲 Freely filtered and then absorbed and performance of the Modification of Diet in Renal Disease
catabolized by renal tubules and Cockcroft-Gault equations for estimating renal func-
䡲 No significant urinary excretion tion. J Am Soc Nephrol 16:763-773, 2005
䡲 Measurement by particle-enhanced neph-
elometric immunoassay (PENIA) with ASSESSMENT OF RENAL PLASMA
high degree of precision and accuracy FLOW (RPF)
䡲 Normal adult values range from 0.54 to
1.55 mg/dL ● Infrequently needed in routine clinical prac-
䡲 Most but not all studies show that serum tice
cystatin C is better index of GFR than ● Derived from rate of clearance of a marker
SCr alone that is totally extracted from plasma after
䡲 Factors that affect serum level of cystatin first pass through kidney; this yields RPF
and that are independent of GFR are still ● Renal blood flow (RBF) can be obtained by
controversial and possibly include older dividing RPF by (1- hematocrit)
age, male sex, smoking, higher weight, ● ␳-aminohippurate (PAH) is most widely
thyroid disease, and higher levels of used marker
C-reactive protein ● PAH clearance gives the effective RPF
䡲 Additional studies, across diverse popu- (ERPF) because part the RBF perfuses a
lations to determine value as index of region that does not contribute to PAH
renal function, still required excretion; PAH clearance is about 10%
lower than actual RPF
ADDITIONAL READING ● Mean values of ERPF are 650 mL/min/
1. Bauer JH, Brooks CS, Burch RN: Clinical appraisal of 1.73 m2 in males and 600 mL/min/1.73 m2
creatinine clearance as a measurement of glomerular filtra- in females
tion rate. Am J Kidney Dis 2:337-346, 1982 ● Other markers that can be used include deter-
2. Levey AS: Nephrology forum: Measurement of renal
function in chronic renal disease. Kidney Int 38:167-184,
mination of plasma clearance of a radioactive
1990 marker such as 131I-hippuran or MAG3
3. Nilsson-Ehle P, Grubb A: New markers for the
determination of GFR: Iohexol clearance and cystatin C ADDITIONAL READING
serum concentration. Kidney Int Suppl 46:S17-S19, 1994 1. Cole BR, Giangiacomo J, Ingelfinger JR, Robson AM:
4. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Measurement of renal function without urine collection: A
Roth D: A more accurate method to estimate glomerular critical evaluation of the constant-infusion technic for determi-
178 ROSNER AND BOLTON

nation of inulin and para-aminohippurate. N Engl J Med releases H⫹ ions and an acid-base
287:1109-1114, 1972 indicator changes color
2. Smith HW, Goldring W, Chasis H: The measurement
of the tubular excretory mass, effective blood flow and
▫ Imprecise and should not be used in
filtration rate in the normal human kidney. J Clin Invest any formal testing
17:263-278, 1938
Assessment of Renal Concentrating Ability

ASSESSMENT OF TUBULAR FUNCTION ● Necessary for patients with polyuria or

hyperosmolality
Concentration and Dilution of Urine: Methods
● Assessed with water deprivation test, which
assess integrated response of pituitary and
● Relies on measurement of urine and plasma kidney
osmolalities: ● Normal water deprivation for 18 to 24
䡲 Osmolality is related to number of par- hours leads to urine osmolality ⬎900
ticles in solution and is independent of mOsm/kg in most healthy persons
charge, size, or density ● Failure to concentrate urine is further as-
䡲 Methods for determination of osmolality sessed with exogenous vasopressin
include: ● Central diabetes insipidus: urine osmolality
E Freezing point depression: extent to increases in response to vasopressin
which freezing point of a solution is ● Nephrogenic diabetes insipidus: no change
depressed below that of distilled water in urine osmolality in response to
is linearly related to osmolality: vasopressin
▫ Advantages: precision and not influ-
enced by excretion of protein or
Assessment of Renal Diluting Capacity
iodinated contrast agents
▫ Disadvantages: expense and labor ● Useful in patients with hypoosmolality
required ● Can be assessed with water loading, but test
E Urine-specific gravity: has low diagnostic yield and is seldom
▫ Measures not only total number of useful in clinical practice
particles but also relative size and ● Defect in diluting process is usually evident
density of particles and thus is not a when serum hypoosmolality coexists with a
measure of urine osmolality urine that is not maximally dilute and does
▫ Refrigeration of urine or excretion not require formal testing
of large amounts of protein, glu-
cose, or contrast agents will result ADDITIONAL READING
in specific gravity increases 1. Sweeney TE, Beuchat CA: Limitations of methods of
▫ Measured with hydrometer or re- osmometry: Measuring the osmolality of biological fluids.
Am J Physiol 264:R469-R480, 1993
fractometer 2. Zerbe RL, Robertson GL: A comparison of plasma
E In general, urine osmolality of 50 vasopressin measurements with a standard indirect test in
mOsm/kg (mmol/kg) is approximately the differential diagnosis of polyuria. N Engl J Med 305:1539-
equivalent to a specific gravity of 1546, 1981
1.000; 300 mOsm/kg, a specific grav-
ity of 1.010; 800 mOsm/kg, a specific Assessment of Urinary Acidification
gravity of 1.020 ● Useful in diagnosis of non-gap metabolic
E Dipstick measurement of urine spe- acidosis: determining extrarenal versus re-
cific gravity: nal causes (renal tubular acidosis) and the
▫ Assumes that as amount of solutes specific renal cause
in urine increases, there will be a ● pH:
concomitant increase in amount of 䡲 Usually measured with reagent test strip
ions (salts) with normal range from 4.5 to 7.8
▫ As concentration of ions increases, 䡲 Should be measured quickly after sample
a polyacid impregnated on the strip is obtained
CORE CURRICULUM IN NEPHROLOGY 179

䡲 Most accurate when obtained using H⫹- 䡲 In healthy subjects, alkalinization of urine
specific electrode on urine collected un- is associated with urine PCO2 approxi-
der oil mately 30 mm Hg greater than blood
䡲 Alkaline pH: urea-splitting organisms, 䡲 Urine values similar to blood PCO2 are
vegetarian diet, diuretics, nasogastric suc- consistent with classical distal renal tubu-
tion, vomiting, alkali therapy lar acidosis
䡲 Acidic pH: metabolic acidosis, high pro- ● Other tests of distal acidification:
tein diet 䡲 Sodium sulfate infusion
● Urine pH itself has little diagnostic informa- 䡲 Response to loop diuretic
tion
● Urine anion gap ([Na⫹ ⫹ K⫹] – Cl⫺)
ADDITIONAL READING
estimates urine NH4⫹ concentration and
renal response to acidosis: 1. Halperin ML, Richardson RM, Bear R, et al: Urine
ammonium: The key to the diagnosis of distal renal tubular
䡲 Positive urine anion gap: decreased acidosis. Nephron 50:1-4, 1980
urine ammonium production (renal tu- 2. Batlle DC: Segmental characterization of defects in
bular acidosis), presence of unmea- collecting tubule acidification. Kidney Int 30:546-554, 1986
sured ketoacids, hippurate, benzoate, 3. Sabatini S, Kurtzman NA: Pathophysiology of the
or penicillin-derivative antibiotics (pi- renal tubular acidoses. Semin Nephrol 11:202-211, 1991
peracillin, ticarcillin)
䡲 Negative urine anion gap: increased urine Assessment of Tubular Function in ARF
ammonium production and extrarenal
● Useful in differential diagnosis of ARF,
source of acidosis
especially in distinguishing prerenal from
● Urine osmolal gap:
renal causes
䡲 Useful when urine anion gap is positive
● Tubular function impaired with acute tubu-
and unclear whether increased excre-
lar necrosis
tion of unmeasured anion is respon-
sible ● Fractional excretion (FE) of sodium, urea,
䡲 Calculation requires measurement of and uric acid allow assessment of tubular
urine osmolality and the urine sodium, function (Table 4):
potassium, urea, glucose 䡲 Fractional excretion of substance x ⫽
䡲 Gap ⫽ measured – calculated urine osmo- [(U/plasma concentration of substance
lality: x)/(U/Scr)] ⫻ 100
E Calculated urinary osmolality ⫽ 2 ⫻ 䡲 Caveats include:
(Na ⫹ K) ⫹ (urea/2.8) ⫹ (glucose/18) E Low FE sodium may be seen early in
E Positive osmolal gap consistent with course of tubular injury accompanying
increased ammonium production rhabdomyolysis, sepsis, administration
● Alkali loading test: test urine pH and of radiocontrast materials, nonoliguric
serum bicarbonate response to bicarbon- forms of ARF, nonsteroidal anti-inflam-
ate infusion (proximal renal tubular acido- matory drug use, acute interstitial nephri-
sis): tis, and with acute glomerulonephritis;
䡲 Infusion of sodium bicarbonate at 0.5 to also may be seen with acute tubular
1.0 mEq/kg/h necrosis in setting of vasoconstrictive
䡲 Urine pH, even if initially acidic, will states such as congestive heart failure
increase rapidly once resorptive thresh- and cirrhosis
old for bicarbonate is exceeded; urine E High FE sodium may be seen in prere-
pH will be ⬎7.5 and fractional excre- nal states in which there is impaired
tion of bicarbonate ⬎15% to 20% as renal tubular reabsorption of sodium
plasma bicarbonate approaches normal such as with diuretic use, bicarbonatu-
● Urine-blood PCO2: ria, glucosuria, recent intravenous con-
䡲 Examined while urine is alkaline trast administration, salt-wasting ne-
180 ROSNER AND BOLTON

phropathy, and mineralocorticoid

ASSESSMENT OF PROTEINURIA
deficiency
● Biomarkers of renal tubular injury are avail-
General Guidelines
able:
䡲 Their current role is more experimental ● Spot urine specimens can be used to detect
than diagnostic and monitor proteinuria
䡲 No current “gold standard” ● First morning specimens are preferred
䡲 Sample biomarkers include brush border method
enzymes such as N-acetyl-␤-glucosamini- ● Screening for proteinuria can be accom-
plished with either standard urine dipsticks
dase, adenosine deaminase–binding pro-
(total proteinuria) or albumin-specific dip-
tein, as well as proteins such as kidney
sticks
injury molecule 1 (KIM-1), neutrophil
● Patients with positive dipstick tests should
gelatinase–associated lipocalin (NGAL), have confirmation with quantitative mea-
and sodium-hydrogen exchanger 3 surement (albumin-creatinine ratio, protein-
(NHE3) creatinine ratio, or 24-hour collection)
● Measurement of low-molecular-weight pro- ● Monitoring proteinuria in patients with
teins that are readily filtered and usually chronic kidney disease should be per-
reabsorbed by proximal tubule: formed with quantitative measurements
䡲 May appear in urine when there is tubu- ● Total protein-creatinine ratio is acceptable
lar injury method if albumin-creatinine ratio is high
䡲 These proteins include ␤2-microglobulin, (⬎500 mg to 1,000 mg/g)
amylase, lysozyme, and retinol-binding pro-
tein Measurement of Total Urine Protein:
䡲 These measurements are not routinely Methodology
useful ● Semiquantitative tests for total urine pro-
tein:
ADDITIONAL READING 䡲 Precipitation tests use 5% sulfosalicylic
1. Carvounis CP, Nisar S, Guro-Razuman S: Significance acid, concentrated nitric acid, or 10%
of the fractional excretion of urea in the differential diagnosis of trichloroacetic acid to precipitate pro-
acute renal failure. Kidney Int 62:2223-2229, 2002 teins in urine; quantity of precipitate is
2. Rabb H: Evaluation of urinary markers in acute renal graded from 0 to ⫹4
failure. Curr Opin Nephrol Hypertens 7:681-686, 1998 䡲 Dipstick test uses a pH indicator dye
3. Miller TR, Anderson RI, Linas SL, et al: Urinary
(tetrabromophenol blue) buffered to pH
diagnostic indices in acute renal failure: A prospective study.
Ann Intern Med 88:47-57, 1978 of 3.0; proteins act to change color of
indicator dye with color varying depend-
ing on concentration of protein present:
Table 4. Urinary Indices in ARF E Detect total protein ⬎10 to 20 mg/dL
䡲 Tests are insensitive but have high speci-
Urinary Study Prerenal Renal
ficity
Specific gravity ⬎1.020 ⬃1.010 䡲 Dipstick false-negative results with im-
Urinary osmolality (mOsm/kg) ⬎500 ⬍300 munoglobulin light chains, tubular pro-
Urine sodium (mEq/L) ⬍20 ⬎20 teins that have positive charges (Table 5)
Fractional sodium excretion (%) ⬍1 ⬎2
Fractional uric acid excretion (%) ⬍7 ⬎15
䡲 False-positive results (Table 5)
Fractional urea excretion ⬍35 ⬎35 䡲 They detect an abnormal concentration,
Low-molecular-weight proteins Low High not an abnormal excretion rate
(␤2-microglobulin, lysozyme) 䡲 Useful for screening purposes
Brush-border enzymes Low High ● Quantitative tests for total urine protein:
(N-acetyl-␤-glucosaminidase)
䡲 24-hour collection:
NOTE. To convert osmolality in mOsm/kg to mmol/kg, E Urinary protein is precipitated with
multiply by 1; sodium in mEq/L to mmol/L, multiply by 1. trichloroacetic or sulfosalicylic acid
CORE CURRICULUM IN NEPHROLOGY 181

Table 5. Causes of False-Positive and False-Negative estimates of daily urine protein excretion
Results in Urinary Measurement of Albumin or and exclude diurnal variations in protein
Total Protein
excretion
False-Positive Results False-Negative Results
Tests for Albumin Excretion: Methodology
Dehydration: increased Excessive hydration: decreased
concentration of concentration of protein in ● Can be quantified in a variety of ways:
protein in urine urine 䡲 Radioimmunoassay (gold standard), im-
Hematuria Other proteins that do not react munoturbimetric method, or laser neph-
Exercise (especially with dipstick (eg, monoclonal
albumin) protein)
elometer are main quantitative methods
Urinary tract infections 䡲 Semiquantitative dipstick measurements
Extremely alkaline also available for detecting microalbu-
urine (pH ⬎ 8) minuria (albumin ⬎30 ␮g/min or 30 to
300 mg/d); use limited to screening only;
ETurbidity is measured with photom- sensitivity, specificity influenced by urine
eter or nephelometer and compared to concentration
a standard ● Appropriate for early detection of renal
E Adequacy of collection ensured by disease and cardiovascular risk
concomitant measurement of urinary ● Albumin-creatinine ratios of single voided
creatinine excretion: specimen; timed collections provide better
▫ Normal value ⬍150 mg/d estimate of albumin excretion
▫ Iodinated contrast material can falsely ● Abnormal values:
elevate the protein concentration 䡲 Microalbuminuria: albumin, 30 to 300
䡲 Single voided specimen: mg/d
E Ratio of protein to creatinine concentra- 䡲 Albuminuria: albumin ⬎300 mg/d
tion in random urine can provide simple ● In established glomerulopathies, there is no
estimate of daily protein excretion evidence that albuminuria rate is more
E Best to obtain serial specimens at informative than total proteinuria
same time of day given circadian
variation of urine protein excretion: Tests for Light Chains: Methodology
▫ Ratio corrects for variations in urine ● Establishes diagnosis of multiple myeloma
concentration
or other monoclonal gammopathy
▫ Kidney Disease Outcomes Quality
● Bence-Jones test: heat:acetic acid precipita-
Initiative recommends first morn-
tion; insensitive and difficult to perform
ing or random spot urine to monitor
● Best test is protein electrophoresis, which
proteinuria
detects monoclonal peak and immunofix-
䡲 Good correlation with values obtained
by 24-hour specimens once creatinine ation to identify specific protein
excretion is known
Spot Versus Timed Urine Collections in
䡲 24-hour specimens or even overnight
collections with calculation of the protein- Assessment of Proteinuria/Albuminuria
creatinine ratio may lead to better true ● See Table 6

Table 6. Comparisons of Methods for Assessment of Proteinuria

Random Urine for Morning Urine for Timed Collections for

Albumin-Creatinine Ratio Albumin-Creatinine Ratio Albumin Excretion

Convenient Convenient Inconvenient

Lower creatinine excretion in women: higher values of albumin-creatinine ratio Not an issue
Lower creatinine excretion in elderly: higher values of albumin-creatinine ratio Not an issue
Greater creatinine excretion in African Americans: lower values of albumin-creatinine ratio Not an issue
182 ROSNER AND BOLTON

● Collections of 24-hour or timed urine speci-

mens are associated with high error rate and
are inconvenient
● Studies comparing spot urine albumin-
creatinine or protein-creatinine ratio with
timed specimens have shown correlation
coefficients ranging from 0.6 to 0.96
● First morning urine specimen is preferred
and shows best correlation with 24-hour
value
● Detailed assessments of precision and bias
in the accuracy of spot urine specimens
versus timed collections is not adequately
known

Interpretation of Proteinuria
● Electrophoretic pattern of urinary proteins
(Fig 1):
䡲 Glomerular proteinuria: albuminuria is
hallmark (making up 60% to 90% of
total proteinuria)
䡲 Tubular proteinuria: low-molecular-
weight proteins predominate with total
urine protein rarely ⬎2 g/d:
E Impaired tubular reabsorption of low-
molecular-weight proteins, or
E Overproduction of low-molecular-
weight proteins, such as light chains
in myeloma
䡲 Selective proteinuria expressed as clear-
ance of IgG over clearance of trans- Fig 1. Electrophoretic patterns of serum and urine
in patients with abnormal protein excretion.
ferrin is reliable indicator of severity
and reversibility of abnormalities of
glomerular proteinuria; patients with
highly selective proteinuria have milder 䡲 Orthostatic:
tubulointerstitial damage, improved E Proteinuria only in erect position with
prognosis, and better response to total proteinuria usually ⬍1 g/d
therapy: E Diagnosis with split 24-hour urine
E Selectivity index (SI) ⫽ urine IgG/ specimen: with 16-hour collection
serum IgG ⫻ serum transferrin/urine while patient upright and 8-hour col-
transferrin lection while recumbent
E SI ⱕ0.10 highly selective; SI ⱖ0.11 E Benign condition
and ⱕ0.20 moderately selective; SI 䡲 Persistent:
ⱖ0.21 nonselective E Almost invariably sign of structural
● Patterns of proteinuria: renal disease and often requires renal
䡲 Intermittent: biopsy for definitive diagnosis
E Due to hemodynamic alterations in
permselectivity ADDITIONAL READING
E Associated with fever, exercise, stress 1. Abuelo JG: Proteinuria: Diagnostic principles and
E Benign prognosis procedures. Ann Intern Med 98:186-196, 1983
CORE CURRICULUM IN NEPHROLOGY 183

2. Weber MH: Urinary protein analysis. J Chromatogr 5. Rodby RA, Rohde RD, Sharon Z, et al: The urine
429:315-344, 1988 protein to creatinine ratio as a predictor of 24-hour urine
3. Ginsberg JM, Chang BS, Matarese RA, Garella S: Use protein excretion in type 1 diabetic patients with nephropa-
of single voided urine samples to estimate quantitative thy. The Collaborative Study Group. Am J Kidney Dis
proteinuria. N Engl J Med 309:1543-1546, 1983 26:904-909, 1995
4. Schwab SJ, Dunn FL, Feinglos MN: Screening for 6. Zelmanovitz T, Gross JL, Oliveira JR, et al: The
microalbuminuria: A comparison of single sample methods receiver operating characteristics curve in the evaluation of a
of collection and techniques of albumin analysis. Diabetes random urine specimen as a screening test for diabetic
Care 15:1581-1584, 1992 nephropathy. Diabetes Care 20:516-519, 1997