Screening for early chronic kidney disease—what method fits best? (original) (raw)

Introduction

Much attention is presently focused on the detection of early chronic kidney disease (CKD). This interest is related to the fact that it is becoming more and more accepted that an impaired kidney function and elevated albuminuria are associated with progressive cardiovascular disease. It is thus important that easy-to-apply and reliable techniques be available to properly define the presence of renal damage. In this comment, we will describe what methods are available to define the presence of CKD, and we will discuss how these methods should be used in daily practice.

Changing clinical nephrology practice over the decades

Nephrology practice has dramatically changed since the early sixties of the previous century. In the early years of nephrology, most attention was directed towards setting up dialysis and transplant programmes. In those years, little attention was paid to the prevention of progressive renal function loss. In the nineties, practice started to change, as it became clear that progressive CKD in subjects with known renal disease (and thus under the attention of the nephrologists) could be slowed down by strict blood pressure control and lowering of proteinuria with Angiotersir Carvering Enzyme (ACE) inhibitors or angiotensin II receptor blockers [ 1 ]. Though in the optimal implementation of such renoprotective regimens much can still be gained, there are indications that the number of patients with end-stage renal disease (ESRD) due to classical renal diseases is diminishing. This favourable sign is, however, overruled by the fact that the number of patients with ESRD due to type 2 diabetes mellitus, hypertension and generalized atherosclerosis is constantly growing, not only in the elderly but also in those under 60 years of age [ 2 ]. Unfortunately, these subjects are often brought to the attention of the nephrologists only at a time when they are close to the need for dialysis, that is at a time when treatment opportunity to prevent a further renal function decline is already limited. As such treatments are well available, these patients should be detected earlier in the course of their progressive CKD.

How can we detect patients not known with a specific renal disease, but at risk for progressive CKD?

As patients with renal failure due to type 2 diabetes, hypertension or generalized vascular disease have in most cases never experienced acute symptoms indicative of renal disease, such as haematuria, severe hypertension or oedema (as patients with glomerular or interstitial diseases), programmes have to be designed to detect them at an earlier phase. The KDOQI guidelines defined the five stages of CKD, dependent on the level of glomerular filtration rate (GFR) and the presence of an elevated urinary albumin excretion, defined as microalbuminuria (30–300 mg albumin/24 h) or macroalbuminuria (>300 mg albumin/24 h) ( Table 1 ) [ 3 ]. The publication of these definitions facilitated the ideas of approaching early CKD.

Table 1.

The five stages of CKD, according to the level of GFR and the presence of an elevated albuminuria. Given are the percentages as found in the PREVEND study in Groningen, The Netherlands (de Zeeuw) and the NHANES study in the USA (Coresh)

Table 1.

The five stages of CKD, according to the level of GFR and the presence of an elevated albuminuria. Given are the percentages as found in the PREVEND study in Groningen, The Netherlands (de Zeeuw) and the NHANES study in the USA (Coresh)

We could detect subjects at risk for progressive CKD and cardiovascular disease by screening for albuminuria. With this approach, we would be able to detect subjects with stages 1 and 2 CKD, who cannot be detected by screening only for GFR. Various review papers recently described the laboratory methods to measure albuminuria, the way urine samples could be collected, the definitions for an abnormally elevated albuminuria, and the way in which a population screening on albuminuria might be organized [ 4–6 ]. It is beyond the scope of this editorial to discuss the pros and cons of these aspects.

The second option is screening for GFR, as patients with stages 3 and 4 CKD may have an impaired GFR also without having micro- or macroalbuminuria. It is clear that accurate GFR measurements using inulin or iothalamate infusions cannot be applied in large-scale screening programmes. Accurate 24 h collections necessary for the calculation of a creatinine clearance are also difficult to apply in such programmes. That is the reason why in the last decade, much attention was focused on the optimal formula to estimate GFR from just one single plasma creatinine measurement and some indices of creatinine production. As the latter is determined by muscle mass of the subject, most formulas use age (the elderly produce less creatinine), sex (women produce less creatinine), race (whites produce less creatinine), and weight or height (leaner and smaller subjects produce less creatinine). The most widely used are the Cockcroft–Gault [ 7 ] and the Modification of Diet in Renan Disease (MDRD) [ 8 ] formula ( Table 2 ).

Table 2.

The Cockcroft–Gault formula and the simplified MDRD formula to estimate glomerular filtration rate

Age is included in years, weight in kilogram and serum creatinine in milligram per decilitre.

Table 2.

The Cockcroft–Gault formula and the simplified MDRD formula to estimate glomerular filtration rate

Age is included in years, weight in kilogram and serum creatinine in milligram per decilitre.

Limitations of estimated GFR measurements

Although GFR estimates are easy to apply, the Cockcroft–Gault and MDRD formula as well all the other published formulas have their limitations. It is only if one is aware of these limitations that a good use of the formulas can be expected.

The creatinine measurement

A major point of concern that affects both formulas is the accuracy of the creatinine assay itself. Calibration of the assay is needed, not only to compare individual laboratory results with each other, but also to standardize the results of an individual over time [ 9 ]. Calibration greatly reduces the bias that is found between estimated GFR and true GFR in many studies. Even after calibration, however, still more than half of the results differ more than 15% from true GFR, and more than one-third differ by more than 30% from the correct value [ 10 ]. Second, serum creatinine is not only dependent on endogenous muscle mass, but also on dietary intake of meat, which thus incorrectly may result in a higher serum creatinine value.

The calculation of estimated GFR

There are fundamental differences between the Cockcroft–Gault and the MDRD formula to estimate GFR. First, the Cockcroft–Gault formula has originally been validated against creatinine clearance as the gold standard, whereas the MDRD formula was developed against iothalamate-measured GFR. As creatinine, but not iothalamate, is excreted not only by filtration but also by secretion, creatinine clearance always exceeds iothalamate clearance. Consequently, the Cockcroft–Gault-based GFR estimates tend to exceed MDRD-based GFR estimates in most subjects. Second, the Cockcroft–Gault formula (which includes weight in the formula) is expressed in millilitres per minute, while the MDRD formula (which does not include weight in the formula) is expressed in millilitres per minute per 1.73 m 2 . This difference makes a direct comparison between the two difficult. In general, clinicians are not used to expressing GFR normalized for standard body surface area.

The bias introduced by the estimates

It has been shown in various studies that MDRD GFR in general underestimates true GFR [ 11–13 ], especially in patients with normal GFR, whereas Cockcroft–Gault GFR overestimates true GFR, especially in patients with impaired kidney function [ 11 ]. The biases of the two formulas may be quite different in selected populations, defined by age, sex, body mass index and also level of GFR [ 14 ]. In epidemiological studies, the impact of not just age [ 15 ] but also of sex, body weight, blood pressure and glucose on renal function will generally result in different conclusions when using an indirect instead of a direct measure of GFR [ 16 ].

The use of fixed cut-off levels for CKD definitions

One should realize that the KDOQI guidelines make their discrimination on stages of GFR based upon fixed cut-off levels, that is a GFR 30–59 (stage 3), or 15–29 ml/min/1.73 m 2 (stage 4). It is, however, wellknown that GFR decreases with advancing age by about 0.8 ml/min/year; it is to be expected that we, by using fixed cut-off levels, will diagnose more elderly subjects to have the worse CKD stages. This may result in unnecessary diagnostics and treatment of the elderly. It will also, and even more unwanted, result in missing this diagnosis in the young and male subjects. This problem could be overcome by making age-specific cut-off values for an impaired GFR. In a recent study in more than 2 million subjects from the Veterans Affairs Health Care System, it was shown that the association of eGFR with mortality was much steeper in younger compared with older subjects. This led the authors to conclude that to properly evaluate the impact of an impaired GFR on mortality, different cut-off values should be used for young compared with older subjects [ 17 ].

An impaired GFR or progressive renal function impairment?

Subjects that reach CKD stage 5 will at some point in time have passed KDOQI stages 3 and 4. Screening for estimated GFR therefore seems logical. However, a point of concern is that diagnosing someone with a GFR below a certain cut-off level, does not necessarily mean that this subject will have a progressive loss of renal function. It may well be that he or she has few nephrons and consequently, a low GFR, that however has been and will be stable for many years. Indeed, a population survey showed that in patients with an increased serum creatinine who were not being treated by renal services, only a minority showed progressive renal function decline during 31 months of follow-up [ 18 ]. We recently showed over 4.2 years of follow-up that the loss of GFR in subjects, selected because of an impaired GFR, was not higher than in the background population (with a normal GFR), while the fall in GFR in subjects with macro-albuminuria, even those with a better preserved GFR, was much more rapid than the background population. Interestingly, cardiovascular prognosis was similarly unfavourable in the subjects with impaired GFR and in the subjects with macroproteinuria [ 19 ]. Thus, it may well be that screening for GFR is not the ideal method to detect patients at risk for progressive renal failure. From a renal perspective, it may be more effective to adopt a screening strategy that is based on the identification of subjects that have the combination of an elevated albuminuria, a decreased GFR and any of the known modifiable progression risk factors, such as hypertension, diabetes, smoking and hyperlipidaemia. This would imply that in analogy to the Framingham risk score for cardiovascular disease, a risk score predicting progressive renal function decline should be applied. Such a renal risk score is however, yet to be developed.

What is the role for the nephrologist?

As it has been shown that ∼5–6% of the population has stages 1 and 2 CKD, and another 5% has a stage 3 CKD [ 20 , 21 ], the workload required to detect and especially to follow these subjects is tremendous. This will not be feasible for nephrologists, and also not advisable, since many of these subjects will not develop stage 5 CKD. The main risk that threatens such subjects is cardiovascular disease. Screening for early CKD thus requires a combined approach by the (cardiovascular) internist, general practitioner, nurse and technician. The nephrologist should take the initiative. The type of health care system in the individual country will dictate the optimal design of the screening programme. The components to consider include: who will do the testing, who will take care of the individual with an abnormal test result, where will the screening take place (e.g. clinic, health fair), and how it will be financially supported.

Conflict of interest statement . None declared.

References

1

Jafar TH, Stark PC, Schmid CH et al . Progression of chronic kidney disease: the role of blood pressure control, proteinuria, and angiotensin-converting enzyme inhibition: a patient-level met-analysis.

Ann Inten Med

2003

;

139

:

244

–252

2

Gansevoort RT, van der Heij B, Stegeman CA et al . Trends in the incidence of treated end-stage renal failure in the Netherlands: hope for the future?.

Kidney Int

2004

;

[Suppl 92]

:

7

–10

3

Kidney Disease Outcome Quality Inititative: K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification and stratification.

Am J Kidn dis

2002

;

39

:

S76

–S92

4

De Jong PE, Gansevoort RT. Screening techniques for detecting chronic kidney disease.

Curr Opin Nephrol

2005

;

14

:

567

–572

5

Gansevoort RT, Verhave JC, Hillege HL et al . The validity of screening based on spot morning urine samples to detect subjects with microalbuminuria in the general population.

Kidney Int

2005

;

94 [Suppl]

:

S28

–S35

6

Levey AS, Eckardt KU, Tsukumato Y et al . Definition and classification of chronic kideny disease: a global position statement from Kidney Disease: Improving Global Outcome (KDIGO).

Kidney Int

2005

;

67

:

2089

–2100

7

Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine.

Nephron

1976

;

16

:

31

–41

8

Levey AS, Greene T, Kusek JW, Beck GJ. Simplified equation to predict glomerular filtration rate from serum creatinine.

J Am Soc Nephrol

2000

;

11

:

828 (A)

9

Coresh J, Astor BC, McQuillan G et al . Calibration and random variation of the serum creatinine assay as critical elements of using equations to estimate glomerular filtration rate.

Am J Kidney Dis

2002

;

39

:

920

–929

10

Hallan S, Asberg A, Lindberg M, Johnsen H. Validation of the MDRD formula for estimating GFR with special emphasis on calibration of the serum creatinine assay.

Am J Kidney Dis

2004

;

44

:

84

–93

11

Lin J, Knight EL, Hogan ML, Singh AK. A comparison of prediction equations for estimating glomerular filtration rate in adults without kidney disease.

J Am Soc Nephrol

2003

;

14

:

2573

–2580

12

Rule AD, Gussak HM, Pond GR et al . Measured and estimated GFR in healthy potential kidney donors.

Am J Kidney Dis

2004

;

43

:

112

–119

13

Poggio ED, Wang X, Greene T et al . Performance of the MDRD and Cockcroft Gault equations in the estimation of GFR in health and in chronic kidney disease.

J Am Soc Nephrol

2005

;

16

:

459

–466

14

Froissart M, Rossert J, Jacquot C et al . Predictive performance of the modification of diet in renal disease and Cockcroft–Gault equations for estimating renal function.

J Am Soc Nephrol

2005

;

16

:

763

–773

15

Verhave JC, Balje-Volkers CP, Hillege HL et al . The reliability of different formulas to predict creatinine clearance.

J Int Med

2003

;

253

:

563

–573

16

Verhave JC, Gansevoort RT, Hillege HL et al . The use of indirect estimates of renal function to evaluate the effect of risk factors on renal function.

J Am Soc Nephrol

2004

;

15

:

1316

–1322

17

O’Hare AM, Bertenthal D, Covinsky KE et al . Mortality risk stratification in chronic kidney disease: one size for all ages?.

J Am Soc Nephrol

2006

;

17

:

846

–853

18

John R, Webb M, Young A, Stevens PE. Unreferred chronic kidney disease: a longitudinal study.

Am J Kidney Dis

2004

;

43

:

825

–835

19

Halbesma N, Kuiken DS, Brantsma AH et al . Macroalbuminuria is a better risk marker than low GFR to identify subjects at risk for accelerated GFR loss in a general population.

J Am Soc Nephrol

2005

;

16

:

327A

20

Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS. Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National health and Nutrition Examination Survey.

Am J Kidney Dis

2003

;

41

:

1

–12

21

De Zeeuw D, Hillege HL, de Jong PE. The kidney, a cardiovascular risk marker and a new target for therapy.

Kidney int

2005

;

[Suppl 98]

:

25

–29

© The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org