American Journal of Kidney Diseases
Volume 42, Issue 4 , Pages 617-622, October 2003

Proteinuria and other markers of chronic kidney disease: a position statement of the national kidney foundation (NKF) and the national institute of diabetes and digestive and kidney diseases (NIDDK) 1

  • Garabed Eknoyan, MD

      Affiliations

    • Department of Medicine, Baylor College of Medicine, Houston, TX, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • Thomas Hostetter, MD

      Affiliations

    • National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • George L. Bakris, MD

      Affiliations

    • Department of Preventive Medicine, Hypertension Training Program, Rush Presbyterian-St Luke’s Medical Center, Chicago, IL, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • Lee Hebert, MD

      Affiliations

    • Department of Internal Medicine, Division of Nephrology, Ohio State University, Columbus, OH, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • Andrew S. Levey, MD

      Affiliations

    • Tufts University School of Medicine, Division of Nephrology, New England Medical Center, Boston, MA, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • Hans-Henrik Parving, MD

      Affiliations

    • Tufts University School of Medicine, Division of Nephrology, New England Medical Center, Boston, MA, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • Michael W. Steffes, MD, PhD

      Affiliations

    • Department of Laboratory Medicine and Pathology, University of Minnesota Medical School, Minneapolis, MN, USA
    • University of Texas Southwestern Medical Center, Dallas, TX, USA
    • ,
  • Robert Toto, MD

      Affiliations

    • Steno Diabetes Center, Gentofte, Denmark
    • University of Texas Southwestern Medical Center, Dallas, TX, USA

Received 10 June 2003; accepted 16 June 2003.

Article Outline

Keywords:  Proteinuria, glomerular filtration rate (GFR), chronic kidney disease (CKD)

 

IN RECENT YEARS, chronic kidney disease (CKD) has become recognized as a major public health problem in the United States. Until the past few years, kidney failure, the last stage of progressive kidney disease, has been the most visible outcome of CKD. The US Renal Data System maintains statistics on treatment of patients with kidney failure by dialysis and transplantation, known as end-stage renal disease (ESRD). The incidence of ESRD has doubled in the United States since 1990. This trend seems likely to continue, albeit at a lower rate, such that the annual incidence of ESRD will increase to 172,000 cases during the next 7 years, and there will be 661,000 individuals receiving ESRD treatment by 2010. A much greater prevalence of earlier stages of CKD has been inferred. Based on data from the Third National Health and Nutrition Examination Survey, 8,000,000 individuals in the United States have significantly decreased kidney function, with an estimated glomerular filtration rate (GFR) less than 60 mL/min/1.73 m2. There are even more individuals with manifestations of kidney damage (particularly albuminuria) without a significant decrease in kidney function. At the current incidence rate of approximately 100,000 new ESRD cases per year, it is evident that most patients with CKD do not progress to ESRD, but likely succumb to cardiovascular disease, which also is the leading cause of death for patients with ESRD on maintenance dialysis therapy.

Poor patient outcomes and the high cost of ESRD care have been the focus of attention heretofore, and, appropriately, much effort has been expended to improve dialysis dose and delivery. However, it is unlikely that technical advances in dialysis can alter significantly the outcomes of existing comorbidities of patients started on maintenance dialysis therapy. To improve dialysis patient outcomes, it will be necessary to improve the health status of patients before they enter a dialysis program. Furthermore, a significant delay in progression and even arrest of CKD now is clinically achievable in many cases. Stated otherwise, patients with CKD should have their disease detected and treated well before the onset of kidney failure and the need for dialysis or transplantation. This now is possible because of increasing evidence that: (1) the adverse outcomes of CKD (kidney failure, cardiovascular disease, and premature death) can be prevented or delayed, (2) earlier stages of CKD can be detected through laboratory testing, (3) treatment of earlier stages of CKD can be effective in reducing progression to kidney failure and preventing systemic complications that develop in the course of progressive loss of kidney function, and (4) treatment of CKD-related cardiovascular risk factors (diabetes, anemia, hypertension, dyslipidemia, and abnormal bone mineral metabolism) at earlier stages of CKD can be effective in reducing the cardiovascular mortality and morbidity of these individuals. These positive outcomes of more effective detection and treatment prompted the National Kidney Foundation (NKF) to expand its Dialysis Outcomes Quality Initiative (DOQI) to encompass the entire spectrum of CKD, and the change of its acronym to K/DOQI for Kidney Disease Outcomes Quality Initiative.

The first set of clinical practice guidelines developed under K/DOQI consisted of “Chronic Kidney Disease: Evaluation, Classification, and Stratification,” published in February 2002.1 These guidelines, developed during a period of 2 years: (1) define CKD and classify its stages independent of the underlying cause, (2) evaluate laboratory measurements for the clinical assessment of kidney function, (3) associate level of kidney function with systemic complications that develop during progressive CKD, and (4) stratify risk for loss of kidney function and development of cardiovascular disease. The guidelines recommend that (1) “All individuals should be assessed, as part of routine health encounters, to determine whether they are at increased risk for developing kidney disease, based on sociodemographic factors”1; (2) individuals identified as being “at increased risk should undergo testing for markers of kidney damage,”1 specifically for proteinuria; and (3) to have their GFR level “estimated from prediction equations that take into account the serum creatinine concentration and some or all of the following variables: age, gender, race, and body size.”1

Prompted by similar public health concerns and as part of its National Kidney Disease Education Program, the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) launched a series of initiatives to provide a stronger scientific basis for the adoption of these measures. As part of this effort and that of the implementation of CKD guidelines, the NIDDK and NKF joined in hosting a conference on “Proteinuria and Other Markers of Chronic Kidney Disease” in Washington, DC, October 8–9, 2002. Participants are listed in the Appendix.

The objectives of the conference were to (1) refine and clarify requirements for the adoption of GFR estimates as a clinically useful marker of kidney function, (2) develop a consensus on the terminology and methods for using proteinuria as a clinical marker of kidney damage, (3) explore the use of proteinuria as a valid end point for clinical trials, and (4) delineate future research needs, especially for markers of kidney damage other than proteinuria. After presentations on these issues by experts in the field, 3 breakout groups addressed these topics and developed the following specific recommendations, which have been reviewed and approved by all conference participants.

I.CLINICAL USE OF GFR ESTIMATES AS A MARKER OF KIDNEY FUNCTION
A.Standardization of Estimated GFR
1.Measurement: Clinical laboratories should measure serum creatinine accurately.
◦Serum creatinine assays should be calibrated to national, and preferably international, reference standards.


2.Laboratory reports: Laboratory reports should caution that:
◦Serum creatinine value alone should not be used to estimate level of kidney function.

◦If serum creatinine level is reported as milligrams per deciliter, values less than 1 mg/dL should be reported to the hundredths, ie, 2 decimal points.


3.Process: Clinical laboratories should use serum creatinine level to compute GFR estimates using a prediction equation.
◦Laboratories should provide GFR estimates as a regular laboratory test using a standard prediction equation.

◦Until a standard equation is agreed on, it is recommended to use the Modification of Diet in Renal Disease (MDRD) equation of 4 variables (serum creatinine level, age, sex, and African American or not) in individuals 18 years or older. The Cockroft-Gault equation, which provides an estimate of creatinine clearance, is less desirable for adults, but preferable to using serum creatinine level alone. The Schwartz or Counahan-Barratt equation should be used for those younger than 18 years.1

◦Prediction equations should be adjusted to account for differences in creatinine calibration between reporting laboratories and that in which the prediction equation was developed by using an international standard.

◦If an international standard is not available, clinical laboratories should calibrate serum creatinine results to standards compatible with those used to develop the prediction equation.

◦Laboratories should request all data needed to compute prediction equations. If requested data cannot be provided, estimates of GFR should be provided for alternative clinical situations. For example, if using the MDRD equation, the result may be reported as the GFR estimate for African Americans and non-African Americans. The clinician ordering the test can then interpret the results as necessary.

◦Laboratory reports of GFR estimates should include appropriate interpretation:
-Give the confidence interval for the estimate.

-List reference values for age and sex.

-Provide an explanation for body surface area adjustment.

-Caution about conditions that affect measurements of serum creatinine, interfering substances, and extremes of body weight.

-Explain indications for timed clearance measurements.

-Facilitate links to CKD definition, CKD guideline action plan, and drug dosing instructions.



Timed urine collections for creatinine clearance measurements are not necessary for routine estimation of GFR. However, they may be helpful in special circumstances, such as cachexia, muscle atrophy, progressive weight loss, or extreme obesity.

B.Education
1.Clinical laboratories should provide instructions to physicians with the introduction of a new laboratory test.

2.There should be widespread provider education about GFR estimates.
◦The NKF and NIDDK should facilitate education at the undergraduate, graduate, and postgraduate (ie, continuing medical education) levels.


3.The public, and especially individuals at increased risk for CKD, should be educated about GFR. The focus should be on individuals with hypertension, diabetes mellitus, a family history of CKD, the elderly, and US ethnic minorities.


C.Research Recommendations
1.Improve prediction equations for estimating GFR using serum creatinine level.

2.Develop new prediction equations using better filtration markers, such as cystatin C.

3.Continue to evaluate the burden of CKD.

4.Evaluate alternative definitions for CKD.



II.ALBUMINURIA MEASUREMENT AND TERMINOLOGY
A.Methods
1.For the diagnosis of CKD in adults and postpubertal children with diabetes, measurement of urinary albumin is preferred to that of total protein. Total protein is more appropriate in children to identify both albuminuria and low-molecular-weight proteinuria.

2.Timed urine collections should not be used. Rather, the ratio of concentrations of urine albumin (in milligrams per deciliter) to urine creatinine (in grams per deciliter) on a spot untimed urine specimen should be used. First-morning spot collections are best for children and adolescents to avoid confounding the effect of orthostatic proteinuria.

3.Laboratories should report albuminuria as milligrams of albumin per gram of creatinine, with 1 reference range of 30 mg albumin/g creatinine or less.

4.Immunoassays for albumin usually have sufficient method sensitivity, and urine creatinine assays are fairly well standardized.

5.At very high levels of proteinuria (spot urine total protein to creatinine ratio > 500 to 1,000 mg/g), measurement of total protein, instead of albumin, on a spot urine sample is acceptable. This should be reported as the ratio of concentrations of total urine protein (in milligrams per deciliter) to urine creatinine (in grams per deciliter), normal range less than 200 mg/g.


B.Process: Use Proteinuria, Albuminuria Risk Assessment, Detection, and Evaluation (PARADE) Recommendations2:
1.Repeat assays at later points in time, at least for patients with diabetes mellitus. Specifically, to identify persistent albuminuria, repeat to confirm values greater than the reference range (≤30 mg albumin/g creatinine) in 2 of 3 tested samples.

2.Patients should refrain from vigorous exercise for 24 hours before sample collection.

3.Refrigerate urine samples for assay the same or next day.

4.One freeze is acceptable, if necessary. Avoid repeated freeze-thaw of specimens. If the local laboratory is not equipped to handle samples, ship overnight on ice.

5.There is no need to acidify or otherwise treat the sample.

6.There is a need for standardization among laboratories for precision and accuracy into the low-normal range of the assay for albumin, ie, 2 mg/L; goal of the interassay coefficients of variation less than 15%.

7.In the future, adjust creatinine value for sex to provide a sex-independent reference range: multiply the concentration in men by 0.68.3


C.Definitions
•Normal is 30 mg albumin/g creatinine or less.

•Microalbuminuria is greater than 30 to 300 mg albumin/g creatinine.

•Macroalbuminuria is greater than 300 mg albumin/g creatinine.



III.ALBUMINURIA AS A CLINICAL MARKER OF KIDNEY DAMAGE
A.Process
1.Populations at increased risk for CKD (ie, those with diabetes mellitus, hypertension, or family history of CKD) should be screened for microalbuminuria, at least annually, as part of their regular health examination.

2.Individuals with documented persistent microalbuminuria (2 of 3 measurements greater than the reference range) who are undergoing treatment for elevated blood pressure, lipid disorders, or both should be retested within 6 months to determine if treatment goals and reduction in microalbuminuria has been achieved.
◦If treatment has resulted in a significant reduction of microalbuminuria, annual testing for microalbuminuria is recommended.

◦If no reduction in microalbuminuria has occurred, blood pressure and lipid levels should be evaluated to determine: (1) if the targets have been achieved, (2) if specific drugs that interfere with the renin-angiotensin-aldosterone system are part of the antihypertensive therapy, and (3) the treatment regimen should be modified accordingly.


3.Recommendations for annual testing of microalbuminuria after its diagnosis and for its evaluation after institution of therapy for hypertension and dyslipidemia deserve to be better established (see Research Recommendations). However, based on the available evidence, continued surveillance of microalbuminuria (items 1 and 2) is recommended to assess progression of CKD and response to therapy.

4.Children should be screened by using a standard dipstick on 2 occasions; once before starting school and then in early adolescence (as recommended by the American Academy of Pediatrics). Subsequent testing should be performed as needed, as recommended in the Pediatric PARADE recommendations.4


B.Research Recommendations
1.The value of testing for albuminuria in various populations at increased risk for CKD, particularly those with hypertension or a family history of CKD or ESRD, should be analyzed.

2.The value of titration of antihypertensive therapy based on changes in albuminuria needs further study.

3.Guidelines for expected reductions in albuminuria with treatment should be developed.

4.New markers of kidney damage should be developed. Among potential markers in urine are specific kidney cells and specific mediators of injury. Patterns of urinary proteins of varying molecular weights or messenger RNA excretion using high-throughput methods may prove valuable.



IV.ALBUMINURIA AS A SURROGATE MARKER FOR CLINICAL TRIALS
A.Accrued data show a strong relationship between level of proteinuria or albuminuria and progression of CKD to kidney failure and fatal cardiovascular disease events. These associations have been useful in devising small-scale pilot and feasibility trials. However, the linkage has not been sufficiently established for use in large multicenter trials. Two exceptions exist. First, trials to prevent conversion from normal urinary albumin excretion (≤30 mg albumin/g creatinine) to microalbuminuria (>30 to 300 mg albumin/g creatinine) in persons with diabetes may be justified because the time to GFR-based declines is long in this condition. Second, trials to treat massive proteinuria (nephrotic syndrome) also may be justified because remission of massive proteinuria itself may yield improved prognosis for kidney and cardiovascular diseases.

B.Research Recommendations
1.The many available databases from previous clinical trials should be assembled and analyzed for relationships of albuminuria to usual and accepted outcome measures (doubling of serum creatinine level, ESRD, cardiovascular disease events, or death). The strength of the relationships should be assessed for the following subgroups and factors:
◦Specific baseline levels of albuminuria or proteinuria

◦Relative and absolute changes in albuminuria or proteinuria with therapy

◦Specific underlying kidney diseases

◦Changes in albuminuria or proteinuria as a composite along with increments in serum creatinine levels less than doubling, eg, 50%

◦Evaluate as an end point the reduction in albuminuria or proteinuria by 50%.


2.Depending on results obtained from analysis of existing databases, consideration should be given for new clinical trials to explore the potential validity of albuminuria as a surrogate marker of progression of CKD.

3.Develop and validate new markers of kidney damage using currently stored samples from completed clinical trials, as well as samples from ongoing clinical trials.



Back to Article Outline

APPENDIX. 

The following individuals were participants in the Proteinuria Conference: Raj Agarwal, MD, Indiana University, Indianapolis, IN; George Bakris, MD, Rush Medical College, Chicago, IL; Stefano Bianchi, MD, Spedali Riuniti Di Livorno, Livorno, Italy; Roland Blantz, MD, University of California, San Diego, CA; Josephine Briggs, MD, NIDDK, National Institutes of Health, Bethesda, MD; Vito Campese, MD, Keck School of Medicine, Los Angeles, CA; Patricia A. Cleary, MS, George Washington University, Rockville, MD; Wayne Comper, PhD, DSc, Monash University, Clayton, VIC, Australia; Josef Coresh, MD, PhD, Johns Hopkins Medical Institutions, Baltimore, MD; Catherine Cowie, PhD, NIDDK, National Institutes of Health, Bethesda, MD; Paul Eggers, PhD, NIDDK, National Institutes of Health, Bethesda, MD; Garabed Eknoyan, MD, Baylor College of Medicine, Houston, TX; Denis Fouque, MD, Universite Claude Bernard, Paris, France; Susan Furth, MD, PhD, Johns Hopkins University, Baltimore, MD; Saul Genuth, MD, Case Western Reserve University, Cleveland, OH; Thomas Greene, PhD, Cleveland Clinic Foundation, Cleveland, OH; Lee Hamm, MD, Tulane University Health Science Center, New Orleans, LA; Lee Hebert, MD, Ohio State University Medical Center, Columbus, OH; Ronald Hogg, MD, Southwest Pediatric Nephrology Study Group, Houston, TX; Thomas Hostetter, MD, NIDDK, National Institutes of Health, Bethesda, MD; Chi-yuan Hsu, MD, MSC, University of California, San Francisco, CA; Bertram Kasiske, MD, Hennepin County Medical Center, Minneapolis, MN; George Kaysen, MD, PhD, University of California, Davis, CA; Paul Kimmel, MD, NIDDK, National Institutes of Health, Bethesda, MD; John Kusek, PhD, NIDDK, National Institutes of Health, Bethesda, MD; James Lash, MD, University of Illinois, Chicago, IL; Kevin Lemley, MD, PhD, Stanford University School of Medicine, Stanford, CA; Andrew Levey, MD, Tufts New England Medical Center, Boston, MA; Nathan Levin, MD, Renal Research Institute, New York, NY; Michael Mauer, MD, University of Minnesota, School of Medicine, Minneapolis, MN; Peter A. McCullough, MD, University of Missouri Kansas City, School of Medicine, Northville, MI; Samy McFarlene, MD, State University of New York Health Science Center of Brooklyn, Brooklyn, NY; Tracy McGowan, MD, Center for Diabetic Kidney Disease and Early Renal Insufficiency Clinic, Philadelphia, PA; Joel Melnick, MD, Abbott Laboratories, Abbott Park, IL; Tim Meyer, MD, Stanford University, Palo Alto, CA; Catherine Meyers, MD, Division of Kidney, Urologic, and Hematologic Diseases, National Institutes of Health, Bethesda, MD; Albert Mimran, MD, Centre Hospitalier Universitaire, Montpelier, France; Marva Moxey-Mims, MD, NIDDK, National Institutes of Health, Bethesda, MD; Robert Nelson, MD, PhD, NIDDK, National Institutes of Health, Phoenix, AZ; Akinlolu Ojo, MD, PhD, University of Michigan Health System, Ann Arbor, MI; Hans-Henrik Parving, MD, Steno Diabetes Center, Gentofte, Denmark; Cloud Paweletz, PhD, Food and Drug Administration, Bethesda, MD; Ronald Portman, MD, University of Texas-Houston Medical School, Houston, TX; Sylvia Paz Ramirez, MD, NKF, Singapore; Giuseppe Remuzzi, MD, Mario Negri Institute for Pharmacological Research, Bergamo, Italy; Luis Ruilope, MD, Carretera de Andalucia, Madrid, Spain; Shahnaz Shahinfar, MD, Merck and Company Inc, Blue Bell, PA; Kumar Sharma, MD, Thomas Jefferson University, Philadelphia, PA; Ashok Singh, PhD, Hektoen Institute for Medical Research, Chicago, IL; Michael Steffes, MD, University of Minnesota, Minneapolis, MN; Manikkam Suthanthiran, MD, Weill Medical College of Cornell University, New York, NY; Doug Throckmorton, MD, Food and Drug Administration, Rockville, MD; Robert Toto, MD, University of Texas Southwestern Medical Center, Dallas, TX; Matthew Weir, MD, University of Maryland School of Medicine, Baltimore, MD; Roger Wiggins, MB, University of Michigan Health System, Ann Arbor, MI; Mark E. Williams, MD, Harvard Medical School, Director of Dialysis, Beth Israel Deaconess Medical Center, Joslin Diabetes Center, Boston, MA.

Back to Article Outline

References 

  1. National Kidney Foundation . K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease (Evaluation, Classification, and Stratification). Am J Kidney Dis. 2002;39(suppl 1):S1–S266
  2. Keane WF, Eknoyan G. Proteinuria, Albuminuria, Risk, Assessment, Detection, Elimination (PARADE) (A position paper of the National Kidney Foundation). Am J Kidney Dis. 1999;33:1004–1010
  3. Jacobs DR, Murtaugh MA, Steffes M, Yu X, Roseman J, Goetz FC. Gender and race-specific determinations of albumin excretion rate using albumin to creatinine ratio in single untimed urine specimens (The CARDIOA study). Am J Epidemiol. 2002;155:1114–1119
  4. Hogg RJ, Portman RJ, Milliner D, Lemley KV, Eddy A, Ingelfinger J. Evaluation and management of proteinuria and nephrotic syndrome in children (Recommendations from a pediatric nephrology panel established at the National Kidney Foundation conference on Proteinuria, Risk, Assessment, Detection, Elimination (PARADE)). Pediatrics. 2000;105:1242–1249
  • 1 Address reprint requests to Kerry Willis, National Kidney Foundation, 30 E 33rd St, New York, NY 10016. E-mail: kerryw@kidney.org

PII: S0272-6386(03)00826-6

doi:10.1016/S0272-6386(03)00826-6

American Journal of Kidney Diseases
Volume 42, Issue 4 , Pages 617-622, October 2003

Article Tools