Renal Failure, 32:153–156, 2010 Copyright © Informa UK Ltd.ISSN: 0886-022X print / 1525-6049 onlineDOI: 10.3109/08860220903491208

153

LRNFCLINICAL STUDY

Quantitative Evaluation of Proteinuria by Estimation of the Protein/Creatinine Ratio in a Random Urine Sample

Comparison of 24 h Proteinuria and Pr/Cr RatioGeorge Kosmadakis, Vasileios Filiopoulos, Christodoulos Georgoulias, Despoina Smirloglou, Theodoros Draganis, and Spiridon MichailDepartment of Nephrology, “Gr. Vosnides,” Laiko General Hospital, Athens, Greece

The aim of this study was to evaluate the severity of pro-teinuria using the protein/creatinine ratio in a random urine sample.In 45 patients (male 28, female 17; mean age 50.68 ± 18.26years) with proteinuria of various causes, we measured the 24-hourprotein excretion per 1.73 m2 of body surface and, during thesame day, the protein/creatinine ratio in three different urinesamples (8 am, 12 pm, 4 pm). The 24 h proteinuria was definedas mild (<1 g), moderate (1–3.4 g), and severe (>3.4 g) in 7, 27,and 11 patients, respectively. The sensitivity for protein/creatinineratio compared to the 24 h proteinuria as a method of referencewas 86–100% in the mild, 78–100% in the moderate, and 73–82%in the severe proteinuria, whereas the specificity was 84–100%,78–83%, and 100% respectively. The patients with better renalfunction had significantly higher proteinuria levels. There was asimilarity in the 24 h proteinuria and the protein/creatinine ratiomeasurements in all renal function and level-of-proteinuriagroups. The protein/creatinine ratio of the morning and middaysamples had a very good association with the 24 h sample,whereas it was not associated significantly with the evening sam-ple (4 pm). In conclusion, the degree of 24 h proteinuria levelscan be evaluated by calculating the protein/creatinine ratio in arandom urine sample collected at any time from morning untilmidday. Protein/creatinine ratio is independent of the severity ofproteinuria or renal function, and it can replace in clinicalpractice the cumbersome 24 h urine collections.

Keywords protein-to-creatinine ratio, proteinuria estimation,methods, 24 h proteinuria, chronic kidney disease

INTRODUCTION

Twenty-four hundred years ago, Hippocrates notedthe association between “bubbles on the surface of theurine” and kidney disease.[1] Proteinuria (PR) evaluation isimportant for the differential diagnosis, follow-up, assess-ment of prognosis, and response to therapy of primaryrenal diseases or systemic disorders with renal involve-ment. Quantitation of urinary protein excretion is used in alarge number of diseases and a wide range of patientgroups. The method most commonly used to measureurinary protein relies on 24-hour urine collections, whichare time-consuming, cumbersome, and often inaccurate.A simplified validated determination of proteinuria isobtained by measuring urine protein/creatinine concentra-tion in a spot urine sample.[2,3] According to the literature,the random urine protein/creatinine ratio (Pr/Cr) providesan accurate assessment of quantitative protein excretionand avoids errors and difficulties associated with 24-hoururine collection.[4] In various studies over the years, asignificant correlation between the Pr/Cr and the 24-hourprotein excretion per 1.73 m2 of body surface (PR/24 h/1.73 m2) has been demonstrated in several conditions, includ-ing diabetes mellitus, pregnancy, and lupus nephritis.[5–7]

In a classic study by Ruggenenti et al., Pr/Cr was foundto be more accurate than the 24-hour urine proteinmeasurement.[8]

Moreover, a circadian rhythm of proteinuria isdemonstrated in patients with glomerular disease, and,consequently, it may be important to define the optimalperiod of the day to obtain a random urine sample.[9]

The aim of the present study was to evaluate theseverity of proteinuria by estimation of the Pr/Cr in a ran-dom urine sample in order to define the impact of renalfunction in the accuracy of proteinuria assessment and toevaluate the optimal period of day for the urine sampleuptake.

Received 10 July 2009; revised 31 July 2009; accepted26 October 2009.

Address correspondence to Dr. George Kosmadakis, 35Metsovou Str, 10683, Athens, Greece; Tel.: (+30) 22410-03300;Fax: (+30) 22410-03301; E-mail: George.Kosmadakis@googlemail.com

Ren

Fai

l Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f C

alif

orni

a Ir

vine

on

11/0

2/14

For

pers

onal

use

onl

y.

154 G. Kosmadakis et al.

MATERIALS AND METHODS

Forty-five patients (28 male, 17 female, mean age50.68 ±18.26 years) with proteinuria of various causeswere studied. The primary causes of proteinuria are listedin Table 1. In the patients of the study, we measured the24 h proteinuria (PR/24 h/1.73 m2) and, during the sameday, the protein/creatinine ratio (Pr/Cr) in samples fromthree different times (8 am, 12 pm, and 4 pm). The studypatients were divided in three groups according to theseverity of proteinuria (see Table 2). The PR/24 h/1.73 m2

was defined as mild (<1 g), moderate (1–3.4 g), or severe(>3.4 g) in 7, 27, and 11 patients, respectively. The studiedpatients were also divided in three groups accordingto their renal function [creatinine clearance (ClCr)]:

>70 mL/min (n = 8), 10–69 mL/min (n = 16), and <10 mL/min (n = 22) (see Table 3). Creatinine clearance was esti-mated using the Cockroft–Gault formula. Turbidimetricmethod was used for the quantitation of proteinuria in thetimed and random urine samples,[10] and the enzymaticmethod was used for the measurement of creatinine con-centration in the random urine sample (RUS).[11]

Statistical Analysis

SPSS for Windows 16.0 (SPSS Inc, Chicago, Illinois,USA) was used in analyzing the data. Normality of distri-butions was checked and found to be fairly Gausian. Wetherefore conducted parametric tests in the statisticalanalysis of the study. We estimated the Pr/Cr sensitivityand specificity according to the severity of proteinuria, thelevels of renal function, and the time of the sampling forthe RUS. Using a multivariate analysis for PR/24 h/1.73m2 and factors of sex, renal function, and method ofestimation of urine protein (24 h, 8 am, 12 pm, 4 pm), weinvestigated the impact of the stratification according tothe PR/24 h/1.73 m2 levels.

We also used a multiple linear regression model for renalfunction with a dependent variable of the PR/24 h/1.73 m2

and independent variable of the Pr/Cr in the different timesin order to show which of the three samples obtained dur-ing the day can predict with better accuracy the PR/24 h/1.73 m2. Results are expressed as mean ± SD, and ap value < 0.05 was considered statistically significant.

Table 1 Primary causes of proteinuria

Glomerulonephritis 21 pts• Vasculitis 5 pts• SLE 4 pts• IgA nephropathy 6 pts• FSGS 3 pts• Membranous nephropathy 1 pt• Minimal change disease 2 pts

Diabetic nephropathy 9 ptsAmyloidosis 3 ptsMultiple myeloma 2 ptsUnknown origin 10 ptsTotal 45 pts

Table 2 Mean levels of PR/24 h 1.73 m2 and P/Cr in three different urine samples according to the level of 24 h proteinuria

Severity of proteinuria

24 h proteinuria levels (g)

PR/24 h × 1.73 m2 (g)

Pr/Cr (8 am) (mg/mg)

Pr/Cr (12 pm) (mg/mg)

Pr/Cr (4 pm) (mg/mg)

Severe >3.5 6.15 ± 4.63 5.95 ± 4.52 5.91 ± 4.50 4.97 ± 4.63Moderate 1–3.4 1.88 ± 0.57 1.87 ± 0.59 1.86 ± 0.59 1.56 ± 0.51Mild <1 0.68 ± 0.24 0.72 ± 0.27 0.68 ± 0.24 0.64 ± 0.22

Table 3 Mean levels of PR/24 h × 1.73 m2 and Pr/Cr in three different urine samples according to renal function [creatinine clearance (ClCr)]

ClCr (ml/min)

PR/24 h × 1.73 m2 (g)

Pr/Cr (8 am) (mg/mg)

Pr/Cr (12 pm) (mg/mg)

Pr/Cr (4 pm) (mg/mg)

>70 3.60 ± 6.64 3.51 ± 6.40 3.44 ± 6.43 3.05 ± 5.8610–69 2.90 ± 2.03 2.77 ± 1.99 2.69 ± 1.94 2.38 ± 2.15<10 2.46 ± 1.62 2.48 ± 1.59 2.52 ± 1.58 2.20 ± 1.52

Ren

Fai

l Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f C

alif

orni

a Ir

vine

on

11/0

2/14

For

pers

onal

use

onl

y.

Comparison of 24 h Proteinuria and Pr/Cr Ratio 155

RESULTS

Using the PR/24 h/1.73 m2 as method of reference, Pr/Crsensitivity was found 86–100% in the mild, 78–100% inthe moderate, and 73–82% in the severe PR/24 h/1.73 m2,whereas the specificity was 84–100%, 78–83%, and100%, respectively. These results confirm the accuracyand reliability of Pr/Cr in the assessment of proteinuria. Wefound no differences on concordance of PR/24 h/1.73 m2

and P/Cr between male and female patients (p = NS).As seen in Table 2, the Pr/Cr ratio was equally

accurate in all levels of proteinuria (p = NS). The samplestaken in the morning and at midday were not significantlydifferent from the PR/24 h/1.73 m2 samples (p = NS) forany level of proteinuria. The samples taken at 4 pm werenot significantly associated with the 24 h urine samples atany proteinuria level (p = 0.021).

In the various groups of patients divided accordingto their renal function (see Table 3) (ClCr: >70 mL/min,10–69 mL/min, and <10 mL/ min), there was no signifi-cant differences between the PR/24 h/1.73 m2 and the Pr/Crtaken at 8 am and at midday (p = NS), nor was thereany association with the evening Pr/Cr measurement(p = 0.018). The levels of Pr/Cr were independent of therenal function of the studied patients (p = NS), whichmeans that the Pr/Cr ratio is equally accurate in any levelsof renal function (p = NS).

DISCUSSION

Proteinuria is one of the most important and accurateindicators of renal injury. The qualitative and semi-quantitative proteinuria measurement in random solitaryurine samples has amplified the ability of the health pro-viders to detect and follow up primary kidney diseases andsystemic conditions associated with proteinuria.[2] Themethod is easy and cheap, and, in many cases, it has beensubstituted for the protein measurements from 24 h urinesamples, which may be cumbersome and inaccurate. Protein/creatinine ratio—in general—accurate corresponds to 24 hproteinuria.[2] In view of the circadian rhythm of pro-teinuria[4] and the potential differences between sexes, weconducted this study in order to measure the accuracy ofurine samples in different hours of the day.

In this study, we found no differences between sexesand a very good concordance between 24 h proteinuria andP/Cr ratio in samples taken at 8 am and 12 pm. Anotherpoint to consider is the possibility of inaccuracies in theresults from the RUS P/Cr measurements due to reduc-tions in urine creatinine excretion in the process of chronickidney disease. PR and Pr/Cr were associated in the differ-ent levels of renal function.

Moreover, it seems that there is a significant discrep-ancy between the protein/creatinine ratio values givenbefore and after midday. The outpatient clinics are veryoften programmed for afternoon hours. Taking the circadianproteinuria fluctuation into consideration, a possible changein the hours of the programmed appointments may causesignificant differences in the expected proteinuria results.This may lead to confusing results concerning the activityof a primary or secondary renal disease and the effective-ness of a given therapy. The results of the present study mayhelp the therapeutic community to critically assess thesedynamic changes and take the appropriate logistical steps inorder to avoid potentially harmful methodological errors.

The results of the study indicate that P/Cr is an accu-rate estimate of quantitative protein excretion. P/Cr in spoturine samples is a precise indicator of proteinuria and areliable predictor of progression of disease in patients withchronic nephropathies, and it represents a simple and inex-pensive procedure in establishing severity of renal diseaseand prognosis. Finally, we concluded that the degree of PRcan be evaluated by calculating the P/Cr in a random urinesample collected at any time from morning until midday.

ACKNOWLEDGMENTS

The authors report no conflicts of interest. The authorsalone are responsible for the content and writing of the paper.

REFERENCES

1. Adams F. The genuine works of Hippocrates. Vol. 2.London: Sydenham Society;1849:766.

2. Price CP, Newall RG, Boyd JC. Use of protein: Creatinineratio measurements on random urine samples for predictionof significant proteinuria: A systematic review. Clin Chem.2005 Sep;51(9):1577–1586.

3. Rodriguez-Thompson D, Lieberman ES. Use of a randomurinary protein-to-creatinine ratio for the diagnosis of signif-icant proteinuria during pregnancy. Am J Obstet Gynecol.2001 Oct;185(4):808–811.

4. Torng S, Rigatto C, Rush DN, et al. The urine protein tocreatinine ratio (P/C) as a predictor of 24-hour urine proteinexcretion in renal transplant patients. Transplantation. 2001Oct 27;72(8):1453–1456.

5. Rodby RA, Rohde RD, Sharon Z, Pohl MA, Bain RP,Lewis EJ. The urine protein to creatinine ratio as a predictorof 24-hour protein excretion in type 1 diabetic patients withnephropathy: The Collaborative Study Group. Am J KidneyDis. 1995;26:904–909.

6. Neithardt AB, Dooley SL, Borensztajn J. Prediction of 24-hour protein excretion in pregnancy with a single voidedurine protein-to-creatinine ratio. Am J Obstet Gynecol. 2002May;186(5):883–886.

Ren

Fai

l Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f C

alif

orni

a Ir

vine

on

11/0

2/14

For

pers

onal

use

onl

y.

156 G. Kosmadakis et al.

7. Leung YY, Szeto CC, Tam LS, et al. Urine protein-to-creatinineratio in an untimed urine collection is a reliable measure ofproteinuria in lupus nephritis. Rheumatology (Oxford). 2007Apr;46(4):649–652.

8. Ruggenenti P, Gaspari F, Perna A, Remuzzi G. Cross sec-tional longitudinal study of spot morning urine protein:Creatinine ratio, 24-hour urine protein excretion rate, glom-erular filtration rate, and end stage renal failure in chronicrenal disease in patients without diabetes. BMJ 1998;316:504–509.

9. Koopman MG, Krediet RT, Koomen GC, et al. Circadianrhythm of proteinuria: Consequences of the use of urinaryprotein: Creatinine ratios. Nephrol Dial Transplant.1989;4(1):9–14.

10. Lizana J, Brito M, Davis MR. Assessment of five quantita-tive methods for determination of total proteins in urine. ClinBiochem. 1977 Apr;10(2):89–93.

11. Panteghini M, IFCC Scientific Division. Enzymatic assaysfor creatinine: Time for action. Clin Chem Lab Med. 2008;46(4):567–572.

Ren

Fai

l Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Uni

vers

ity o

f C

alif

orni

a Ir

vine

on

11/0

2/14

For

pers

onal

use

onl

y.