Chronic renal failure

Introduction

The measurement of proteinuria may help to establish a diagnosis and predict the outcome of most renal diseases[1 2] but requires the measurement of concentration in timed urine collections. Twenty four hour collections are commonly used to smooth the wide fluctuations in proteinuria over the day[3] but are time consuming and often imprecise. An alternative approach avoiding timed urine collections is the measurement of the protein:creatinine ratio in single random urine specimens.[4-8] In insulin dependent diabetic patients with nephropathy the protein:creatinine ratio correlates with 24 hour urinary protein excretion rate.[9] Whether the ratio also correlates. with the excretion rate in non-diabetic patients with renal disease, however, has not been established. An additional unresolved question in both diabetic and non-diabetic patients with chronic nephropathies is whether the ratio, by preventing the errors related to imprecise urine collections, may predict the progression of chronic renal disease over time even more reliably than 24 hour proteinuria.

We used a cross sectional design to determine whether the protein:creatinine ratio is a reliable indicator of 24 hour urinary protein excretion rate in non-diabetic patients with proteinuria and chronic renal disease. We then investigated and compared the reliability of this ratio and 24 hour proteinuria in predicting the progression of renal disease in the long term.

Patients and methods

Patients

One hundred and seventy seven patients with non-diabetic chronic renal disease and persistent clinical proteinuria were referred to this clinic for evaluation of their eligibility to a multicentre clinical trial to evaluate the efficacy of ramipril in the treatment of nephropathy.[10 11] All these patients provided spot morning and timed urine collections and entered the cross sectional phase of the study. Of the 177 screened patients, 98 satisfied the inclusion criteria of the study (creatinine clearance 20-70 ml/min/1.73 [m.sup.2] and urinary protein excretion rate persistently [is greater than] 1 g/24 hours for at least 3 months with no evidence of overt heart failure or urinary tract infection) and were enrolled in the ramipril trial. All these patients were therefore followed prospectively and entered the longitudinal phase of the present study. The main baseline characteristics of the 98 patients considered here are shown in table 1 and did not differ significantly from those of the whole population of the trial.[11]

[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]

Urine specimens

A 24 hour urine collection was undertaken by 177 outpatients the day before the scheduled clinic visit. All the subjects were instructed to begin the 24 hour collection immediately after completion of the first voiding in the morning and to collect all urine for 24 hours, including the final void at the completion of the 24 hour period. Specimens from 24 hour urine collections and from untimed morning urine collections were then obtained in the clinic for the measurement of protein and creatinine concentration.

Follow up

The 98 patients entering the ramipril trial were a priori stratified for baseline urinary protein excretion rate [is less than] 3 or [is greater than or equal to] 3 g/24 hours[10 11] and were then randomly allocated (on a 1:1 basis) to 5 year treatment with the angiotensin converting enzyme inhibitor ramipril (1.25 to 5.0 mg/day) or with matched placebo plus an alternative treatment as deemed appropriate to achieve and maintain diastolic blood pressure [is less than] 90 min Hg.[10 11] Each patient was examined by a physician at baseline, every month during the first 3 months, and every 3 months thereafter. At each examination blood pressure and heart rate were measured with the patient in the sitting position in the morning and before the ingestion of the study drugs. Serum creatinine and electrolyte concentrations were assessed as were other serum biochemicals (uric acid, glucose, cholesterol, triglycerides, liver enzymes, and bilirubin). A complete blood count was done and 24 hour urine protein, sodium, and urea excretion measured. At baseline and at 1, 3, and 6 months after randomisation then every 6 months the glomerular filtration rate was determined centrally, at the Mario Negri Institute, by the plasma clearance of non-radioactive iohexol.[12] Serum creatinine concentration was measured every 3 months by standard procedures.

Laboratory methods

Urine specimens for creatinine and protein measurements were frozen at - 20[degrees]C until analysis was performed. Creatinine concentration (mg/dl) was determined on a Beckman Creatinine Analyzer II (Brea, California) with the modified Jaffe rate method. Protein concentration (mg/dl) was determined with a Synchron CX5 Beckman Analyzer. The urine protein:creatinine ratio was obtained by dividing the urinary protein concentration by the urine creatinine concentration (as this results in a ratio rather than an absolute number SI units have not been used).

Statistical analysis

To evaluate the relation between spot morning urine in all the 177 screened patients a single slope linear model determined by the least squares method was used. A logarithmic transformation of the data was done before linear regression analysis to correct the non-constant variability of the observed points around the regression line.

The 98 patients entering the ramipril trial who had at least three measurements of glomerular filtration rate (including baseline) were divided a posterior into three groups with lowest, middle, or highest baseline values of spot morning urine protein:creatinine ratio. Thus, two cut off values of 1.7 and 2.7 were identified that segregated the third of patients (33 of 98) with the lowest and the highest ratio values, respectively, from the remaining 32 patients who had ratio values of 1.7 to 2.7 and were therefore considered in the middle third. Baseline data for the three groups were compared with Wilcoxon's test. Univariate correlation analysis between continuous baseline variables--including protein:creatinine ratio in spot morning urine samples and 24 hour urinary protein excretion rate--and slopes of decline of glomerular filtration rate was carried out with the Pearson correlation coefficient (r).[13] Univariate correlation between dichotomous variables and the rate of decline of the glomerular filtration rate was carried out with the point biserial correlation coefficient.[14] Multivariate analysis was done with multiple linear regression.[15] Progression to end stage renal failure was analysed with univariate and multivariate analysis by using the Cox proportional hazards model.[16] Because of their skewed distribution, protein:creatinine ratio, 24 hour urinary protein excretion rate, and serum cholesterol and triglyceride concentrations were log transformed before analysis. Data analysis was done with the SAS package.[17] Data were expressed as mean (SD) unless otherwise stated. Significance was set at 0.05 (two tailed)

Results

Protein:creatinine ratio and 24 hour urinary protein excretion rate

The correlation between protein: creatinine ratio values in spot morning urine specimens and 24 urinary protein excretion was highly significant (P=0.0001), although the correlation decreased with increases in the ratio. Log-log transformation of the data allowed linear regression analysis (r = 0.932, P [is less than] 0.0001). As the regression line and the line of unity (fig 1) are almost identical, an estimate of 24 urinary protein excretion in grams per day can be made from a direct translation of the random protein:creatinine ratio--that is, 24 hour urinary protein = urinary protein ratio.

[Fig 1 ILLUSTRATION OMITTED]

Urinary protein ratio, decline in glomerular filtration rate, and progression to end stage renal failure

Baseline clinical and laboratory measurements according to thirds of the three spot morning urine protein:creatinine ratios are given in table 1. The prevalence of glomerular and non-glomerular diseases was comparable in the three subgroups. Blood pressure, and serum cholesterol concentrations were higher in the highest and middle compared with the lowest third. Distribution to the two study treatments was comparable in the overall study population and within each third because patients were stratified for baseline urinary protein excretion rate before randomisation. During the whole study period, mean (SD) diastolic blood pressure was comparable in the three groups (lowest 87.8 (10.8) mm Hg; middle 90.3 (10.9) mm Hg; highest 91.6 (9.4) mm Hg).

After linearity assessment the individual slopes were found adequate to describe the rate of decline in the glomerular filtration rate. The actual mean rate of decline in the whole cohort of 98 patients was -- 0.46 (1.34) ml/min/1.73 [m.sup.2]/month. The rate of decline significantly correlated (P [is less than] 0.0005) with baseline protein:creatinine ratios (fig 2). Of interest, when data on mean decline were analysed according to thirds of baseline urinary protein:creatinine ratios it emerged that protein:creatinine ratio segregated three different populations of patients with predictably different rates of decline in glomerular filtration rate (lowest (n=33): protein:creatinine ratio [is less than] 1.7, decline in glomerular filtration rate -0.31 (0.21) ml/min/[1.73m.sup.2]/month; middle (n=32): ratio 1.7-2.7, decline -0.48 (0.22) ml/min/[1.73m.sup.2]/month; highest (n=33): ratio [is greater than] 2.7, decline -- 0.90 (0.37) ml/min/[1.73m.sup.2]/month (P [is less than] 0.05 v lowest)).

[Fig 2 ILLUSTRATION OMITTED]

Univariate correlation analyses between baseline variables listed in table 1 and rate of decline of glomerular filtration rate showed that blood pressure, protein:creatinine ratio, and 24 hour urinary protein were significantly associated with a faster decline, with the ratio being the strongest predictor (table 2). On multivariate analysis the ratio was the only variable significantly predictive of a faster decline.

Table 2 Results of univariate and multivariate analyses of correlation between baseline variables and rate of decline of glomerular filtration rate

Funding: Hoechst Marion Roussel supported the ramipril efficacy in nephropathy (REIN) trial.

Conflict of interest: None.

[1] Remuzzi G, Bertani T, Is glomerulosclerosis a consequence of altered glomerular permeability to macromolecules? Kidney Int 1990;38:384-94.

[2] Rossing P, Hommel E, Smidt UM, Parving H. Impact of arterial blood pressure and albuminuria on the progression of diabetic nephropathy in IDDM patients. Diabetes 1993;42:715-9.

[3] Kassirer JP, Harrington JT. Laboratory evaluation of renal function. In: Schrier RW, Gottschalk CW eds. Diseases of the kidney IV. Vol 1. Boston: Little, Brown, 1988: 393-491.

[4] Ginsberg JM, Chang BS, Matarese RA, Garella S. Use of single void urine samples to estimate quantitative proteinuria. N Engl J Med 1983; 309:1543-6.

[5] Lemann J Jr, Doumas BT. Proteinuria in health and disease assessed by measuring the urinary protein:creatinine ratio, Clin Chem 1987;33:297-9.

[6] Schwab SJ, Christensen RI, Dougherty K, Klahr S. Quantitation of proteinuria by the use of protein to creatinine ratio in single urine samples. Arch Intern Med 1987; 147:943-4.

[7] Warram JH, Gearin G, Laffel L, Krolewski AS. Effect of duration of type I diabetes on the prevalences of stages of diabetic nephropathy defined by albumin:creatinine ratio (A:C ratio). J Am Soc Nephrol 1996;7:930-7.

[8] Shield JP, Hunt LP, Batum JD, Pennock CA. Screening for diabetic microalbuminuria in routine clinical care: which method? Arch Dis Child 1995;72:524-5.

[9] Rodby RA, Rohde RD, Sharon Z, Pold MA, Bain RP, Lewis EJ, for the Collaborative Study Group, The urine protein to creatinine ratio as a predictor of 24-hour urine protein excretion in type 1 diabetic patients with nephropathy. Am J Kidney Dis 1995;26:904-9.

[10] Gruppo Italiano, Studi Epidentiologici in Nefrologia (GISEN). A long-term, randomized clinical trial to evaluate the effects of ramipril on the evolution of renal function in chronic nephropathies. J Nephrol 199 1;3:193-202.

[11] GISEN Group (Gruppo Italiano di Studi Epidemiologici in Nefrologia). Randomized placebo-controlled trial of affect of ramipril on decline in glomerular filtration rate and risk of terminal renal failure in proteinuric, non-diabetic nephropathy. Lancet 1997;349:1857-63.

[12] Gaspari F, Perico N, Ruggenenti P, Mosconi L, Amuchastegui CS, Guerini E, et al. Plasma clearance of non-radioactive iohexol as a measure of glomerular filtration rate in patients with normal or impaired renal function. J Am Soc Nephrol 1995;6:257-63.

[13] Draper N, Smith H. Applied regression analysis. 2nd ed. New York: Wiley, 1981.

[14] Kendall KG, Stuart A. The advanced theory of statistics Vol 2. Inference and relationship. 3rd ed. London: Griffin, 1973:323-4.

[15] Gordge MP, Leaker BR, Rylance PB, Neild GM. Haemostatic activation and proteinuria as factors in the progression of chronic renal failure. Nephrol Dial Transplant 1991;6:21-6.

[16] Mc Cullagh P, Nelder JA. Generalized linear models. 2nd ed. New York: Chapman Hall, 1989.

[17] SAS Institute. SAS/STAT user's guide. Version 6. 4th ed. Vol 1 and 2. Cary, North Carolina: SAS Institute, 1989.

[18] Eddy AA, McCulloch L. Liu E, Adams J. A relationship between proteinuria and acute tubulointerstitial disease in rats with experimental nephrotic syndrome. Am J Pathol 1991; 138:1111-23.

[19] Bertani T, Cutillo F. Zoja C, Broggini M, Remuzzi G. Tubulointerstitial lesions mediate renal damage in adriamycin glomerulopathy. Kidney Int 1986;30:488-96.

[20] Remuzzi G. Abnormal protein traffic through the glomerular barrier induces proximal tubular cell dysfunction and causes renal injury. Curr Opin Nephrol Hypertens 1986;4:339-42.

[21] Remuzzi G, Ruggenenti P, Benigni A. Understanding the nature of renal disease progression. Kidney Int 1997;51:2-15.

(Accepted 27 November 1997)

RELATED ARTICLE: Key messages

* The protein:creatinine ratio measured in spot morning urine samples is a simple and reliable indicator of 24 hour urinary protein excretion rate and can therefore be used to quantify proteinuria without the need for timed urine collection

* Spot morning urinary protein:creatinine ratio is the strongest baseline predictor of progression of renal disease in non-diabetic patients with chronic nephropathies

* Compared with 24 hour urinary protein excretion rate, the spot morning ratio is an even more reliable predictor of decline in glomerular filtration rate and progression to end stage renal failure and represents a simple and inexpensive procedure in the determination of severity of renal disease and prognosis

Mario Negri Institute for Pharmacological Research, Clinical Research Centre for Rare Diseases, Via Gavazzeni 11, 24125 Bergamo, Italy

Piero Ruggenenti, doctor

Flavio Gaspari, chemist

Annalisa Perna, statistical scientist

Giuseppe Remuzzi, director

Correspondence to: Dr Ruggenenti ruggenenti@irfmn. mnegri.it

BMJ 1998;316-504-9

COPYRIGHT 1998 British Medical Association
COPYRIGHT 2000 Gale Group