NDT Advance Access published online on June 20, 2008
Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfn331
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CKD—fiction not fact
Correspondence and offprint requests to: E-mail: glassock{at}cox.netSir,
We write in response to the editorial by Coresh et al. [1] and the commentary by De Jong and Gansevoort [2] on the question ‘The Epidemic of Chronic Kidney Disease: Fact or Fiction?’ posed in our editorial [3] and published simultaneously.
The central question of this debate is: ‘Should the diagnosis of chronic kidney disease be based on thresholds of estimated GFR (eGFR) that do not take into account the decline in actual GFR with ageing and the effects of gender on the normal levels of both actual and estimated GFR’? Our answer is an unequivocal ‘No’.
A subsidiary and related question addressed by de Jong and Gansevoort concerns the role of assessing abnormal albuminuria in identifying patients with true ‘chronic kidney disease’. We acknowledge this opinion and agree that such measurements do have a role, but we suggest that further research is needed to establish whether the patho-physiological connection of albuminuria is with ‘kidney’ or with ‘vascular’ disease. These are serious questions and the debate is not a ‘squabble’ (a petty, noisy discussion of a trivial issue). The outcome of this debate will affect millions of individuals throughout the world and directly impinges on our ability to quantify accurately the overall global burden of ‘chronic kidney disease’. The public health implications of this debate are enormous. We submit that without any correction for the effects of age and gender on eGFR thresholds for defining ‘chronic kidney disease’, this burden will be grossly overestimated, particularly among the elderly and among females with what is currently defined as ‘Stage 3 Chronic Kidney Disease’ by both KDOQI and KDIGO criteria. We applaud the comment of de Jong and Gansevoort [2] that to label such patients as having CKD when abnormal albuminuria is absent is no longer appropriate. We believe that when an appropriate and meaningful restructuring of the CKD classification schemata is agreed upon, then estimates of the prevalence of CKD will be both realistic and credible. Our examination of the available data leads us to conclude that CKD is endemic and not epidemic in the population at large. Whether it is ‘common’ or not depends on one's perspective. Treated ESRD (those receiving renal replacement therapy, RRT) is not common (
0.17% of the US population and 0.07% of the UK population) and according to the most recent report of the USRDS, the incidence of treated ESRD is stable or declining in the USA and in 22 of 29 countries in which such incidence data are tracked by the USRDS [4]. This hardly conforms to the definition of an ‘epidemic’. Because of the aforementioned flaws in the proper identification of legitimate CKD in population-wide surveys using current classification methods, we believe that the estimates of 13.1% for total Stage 1–4 CKD and 7.7% for Stage 3 CKD alone [5] are spurious. This is largely because they include a sizeable number of elderly patients (particularly females) who have ‘normal’ eGFRs adjusted for age and gender and no other manifestations of kidney injury, including albuminuria. Identifying such individuals as suffering CKD does them a disservice and exaggerates the burden of CKD in the community-at-large. Based on limited available data we estimate that if a threshold of eGFR below the 5th percentile adjusted for age and gender and evidence for kidney injury (e.g. abnormal albuminuria) were included as criteria for the definition of Stage 3 CKD, the estimated prevalence rate of Stage 1–4 CKD would decrease by 40% or more, to values closer to <7%, primarily due to a reduction in the prevalence of Stage 3 CKD. Studies using new criteria for defining CKD and its application to existing databases are needed. We also question whether the prevalence of CKD in the community-at-large is increasing at a rate out of proportion to that expected by secular trends in the population as a whole.
Thus, we believe that CKD is not as common as asserted and is ‘endemic’ rather than ‘epidemic’ in character. The commentary of de Jong and Gansevoort focuses primarily on albuminuria rather than eGFR. We would question whether ‘micro-albuminuria’ in the absence of a decline in eGFR to abnormal levels (e.g. Stage 1 and 2 CKD by current criteria) should be regarded as a manifestation of ‘chronic kidney disease’ rather than a reflection of a generalized vascular disorder (‘chronic vascular disease’) in which the kidneys, along with other organs, are the victim rather than the culprit. Further, the hypothesis that ‘microalbuminuria’ is a ‘window into the vascular system’ is not proven, as has been pointed out by Couser [6]. It is our duty to stay our hand in promoting public health initiatives (such as population-wide screening for CKD using eGFR and/or measurements of urinary albumin excretion) until there is a sound patho-physiological justification for proceeding with such a massive undertaking.
The issue of whether CKD constitutes an independent ‘risk’ factor for cardiovascular disease (CVD) or vice versa remains unresolved. The epidemiological associations do not thus far define what is cause and what is effect. Furthermore, individuals with eGFR <60 ml/min/1.73 m2 without any abnormal albuminuria (currently also defined as Stage 3 CKD), but who still have eGFR values above the 5th percentile adjusted for age and gender, have not clearly been shown to have any greatly increased risk of fatal of non-fatal cardiovascular events (i.e. >20% above the rate observed in those with eGFR 
60 ml/min/1.73 m2 [7]. Adjustment for all of the traditional cardiovascular risk factors, such as those included in the Framingham Risk score, attenuates the risk relationship of CKD with CVD, although such scoring systems do still tend to underestimate the risk of events in subjects with CKD Stage 3, at least as currently defined [8]. It has also been shown that traditional risk factors have a more powerful influence on CVD risk than non-traditional risk factors in those subjects with an eGFR of <60 ml/min/1.73 m2 [9], but the effect of the presence or absence of proteinuria has not been examined.
The current classification schema also raises ethical issue in transplantation, since an otherwise healthy 60-year-old donor of a kidney would almost certainly be classified as having Stage 3 CKD post-donation [10]. According to current views, one would need to inform the donor that by allowing his or her eGFR to be surgically reduced, to levels currently defined as CKD Stage 3, he or she will be exposed to an additional risk of developing CVD later in life. This is unproven and unlikely.
Finally, the current classification system leads to absurdities in the recommended approach to patients thought to have CKD. Is it appropriate to label both a 30-year-old man with an eGFR of 55 ml/min/1.73 m2 and 3+ proteinuria and a 60-year-old woman with an identical eGFR but no proteinuria as having the same ‘Stage’ of a postulated ‘disease’? We think not.
Classifying patients inappropriately as having CKD when they do not has many implications. Evaluation, testing and referral should be based on conventional criteria for CKD, and we should desist from relying on eGFR alone. Until the methods of assessing eGFR are prospectively verified in diverse populations (e.g. varying in ancestry, body habitus, diet), we should restrain efforts to mandate reporting of calculated eGFR routinely whenever a serum creatinine level is randomly assessed, but encourage cognitive evaluation of the serum creatinine concentrations (and calculated eGFR) within the clinical context of individual subjects. At present we would not describe subjects with a reduced eGFR as ‘healthy’ nor would we describe them as necessarily ‘diseased’.
We are certain that the purported ‘epidemic’ of CKD, as currently defined, is a fiction not a fact.
Conflict of interest statement. None declared.
1 Geffen School of Medicine at UCLA Los Angeles CA, USA 2 Oxford Kidney Unit Oxford, UK
References
- Coresh J, Stevens LA, Levey AS. Chronic kidney disease is common: what do we do next. Nephrol Dial Transplant (2008) 23:1122–1125.
[Free Full Text] - De Jong PE, Gansevoort R. Fact or fiction of the epidemic of chronic kidney disease—let us not squabble about estimated GFR only, but also focus on albuminuria. Nephrol Dial Transplant (2008) 23:1092–1095.
[Free Full Text] - Glassock RJ, Winearls C. An epidemic of chronic kidney disease: fact or fiction? Nephrol Dial Transplant (2008) 23:1117–1121.
[Free Full Text] - United States Renal Data System. 2007 Annual Report. Atlas of chronic kidney disease and end-stage renal disease in the United States. Am J Kidney Dis (2008) 51(Suppl_1):s244.
- Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidney disease in the US. JAMA (2007) 298:2038–2047.
[Abstract/Free Full Text] - Couser WG. Chronic kidney disease: the promise and the perils. J Am Soc Nephrol (2007) 18:2803–2805.
[Free Full Text] - Go AS, Chertow GM, Fan D, et al. Chronic kidney disease and the risks of death, cardiovascular; events and hospitalization. N Engl J Med (2004) 351:1296–1305.
[Abstract/Free Full Text] - Weiner DE, Tighiourat H, Elsayed EP, et al. The framingham predictive instrument in chronic kidney disease. J Am Coll Cardiol (2007) 50:217–224.
[Abstract/Free Full Text] - Shlipak MG, Fried LF, Cushman M, et al. Cardiovascular mortality risk in chronic kidney disease: comparison of traditional and novel risk factors. JAMA (2005) 293:1737–1745.
[Abstract/Free Full Text] - Wan RK, Spalding E, Winch D, et al. Reduced kidney function in living kidney donors. Kidney Int (2007) 71:1077.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||