letter to the editor
University, Xi’an, Shaanxi, China and 3Department of Geriatrics, Xijing Hospital, Fourth Military Medical University, Xi’an, Shaanxi, China Correspondence: Daiming Fan, State Key Laboratory of Cancer Biology, Department of Nephrology, Xijing Hospital, Fourth Military Medical University, Xi’an 710032, China. E-mail: [email protected]
In our cohort of type I diabetic patients with normal renal function without proteinuria, we could not confirm the influence of diabetes on the relationship between CysC and GFR. The effect, as found by Stevens et al.,1 may be due to the interaction with proteinuria, which was present in the majority of their patients.
Cystatin C levels are unaltered in patients with diabetes mellitus and normal renal function
Kidney International (2009) 76, 462; doi:10.1038/ki.2009.167
To the Editor: In their recent article, Stevens et al.1 have reported that factors other than glomerular filtration rate (GFR) affect serum cystatin C (CysC) levels. Specifically, they mentioned the impact of diabetes mellitus, which was associated with 8.5% higher levels of CysC. However, proteinuria was also associated with CysC and, as admitted by the authors, diabetes and proteinuria were strongly associated in their patient cohorts. Thus, their data did not allow for determining the independent effects of diabetes mellitus or proteinuria. We have previously reported GFR data of a cohort of patients with normo-albuminuric diabetes type I and age- and sex-matched controls.2 From this cohort, we have selected 33 patients and 38 controls with matching GFR (inulin clearance) and evaluated the relationship between CysC and GFR. At baseline, the mean GFR was 112±9 ml/min per 1.73 m2 and mean CysC was 0.68±0.1 mg/l. In univariable regression analysis, age (P ¼ 0.016), gender (P ¼ 0.001), serum albumin (P ¼ 0.001), and serum urea (P ¼ 0.002) significantly influenced the relationship of CysC with GFR. Diabetes was not a predictor of this relationship (Figure 1). In multivariable analysis age (P ¼ 0.003), gender (P ¼ 0.007), and serum urea (P ¼ 0.004) remained independent predictors.
Cystatin C (mg/l)
GFR (ml/min per 1.73 m2)
Figure 1 | Relationship between Cystatin C levels and glomerular filtration rate (inulin clearance). Diabetes does not influence this relationship. 462
Stevens LA, Schmid CH, Greene T et al. Factors other than glomerular filtration rate affect serum cystatin C levels. Kidney Int 2009; 75: 652–660. Vervoort G, Willems JL, Wetzels JFM. Assessment of glomerular filtration rate in healthy subjects and normoalbuminuric diabetic patients: validity of a new (MDRD) prediction equation. Nephrol Dial Transplant 2002; 17: 1909–1913.
Julia M. Hofstra1, Gerald Vervoort1, Johannes L. Willems2 and Jack F.M. Wetzels1 1
Department of Nephrology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands and 2Department of Clinical Chemistry, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands Correspondence: Julia M. Hofstra, Department of Nephrology, Radboud University Nijmegen Medical Center, PO-box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: [email protected]
Dividing stage 3 of chronic kidney disease (CKD): 3A and 3B Kidney International (2009) 76, 462–463; doi:10.1038/ki.2009.178
To the Editor: Changing the chronic kidney disease (CKD) staging system, as proposed by Winearls and Glassock,1 could have a negative impact on prevention. One of the hardest tasks faced by a nephrologist is breaking the news about imminent dialysis to an unsuspecting patient. This gets ever more painful if early signs of kidney disease, such as a mildly depressed glomerular filtration rate (eGFR), were somehow uncovered in the past but the patient was reassured without further testing and sent home only to learn otherwise at a later date. Bad as it seems, this is a recurrent experience for any nephrologist. Therefore, we were concerned with this proposition to revise the Kidney Disease Outcomes Quality Initiative (KDOQI) staging system of CKD and eliminate the current early stages of the system. Although we recognize that a proportion of healthy individuals may be transitorily mislabeled as CKD patients, we believe that it is better to raise the possibility of incipient CKD and discard it afterward than to miss diagnosis by not looking for it. The current staging system of CKD has been very successful in bringing worldwide uniformity to the nomenclature and has met the goal of raising awareness within the renal and nonrenal medical communities to new levels. Rather than a matter of opinion, this is a ripe area for scientific inquiry. We challenge the renal community, as would be the case with any new classification system, to use tools of diagnostic statistics and to determine the predictive values of KDOQI staging within the different categories so that we can act and inform our patients appropriately. However, we do not disagree that a minor review of CKD staging may be in order. The wide range of GFR that defines Kidney International (2009) 76, 459–464
letter to the editor
stage 3 (30–59 ml/min per 1.73 m2) implies that a person with a GFR of 55 ml/min per 1.73 m2 would have a potential for complications similar to that of someone with a GFR of 35 ml/min. For example, it has recently been shown that patients with CKD and GFR o45 ml/min per 1.73 m2, particularly the older patients, experience faster disease progression.2 We suggest that it would be clinically sound to subdivide stage 3 into stages 3A (45–59 ml/min) and 3B (30–44 ml/min), as these two ranges may be associated with different clinical patterns and risks. Patients with stage 3B should probably be referred earlier to specialized renal care. Finally, the classification proposed by the authors using percentiles to determine ranges of renal function is not easily applicable in clinical practice. We contend that a change in a system successfully used at this moment may cause disorientation in the medical community and undermine our credibility. Present staging has preventive and epidemiological roles; additional tools can be used and developed specifically for research aims, but the contribution to nephrology practice of the current staging system and the widespread use of eGFR formulae should not be reverted. 1. Winearls CG, Glassock RJ. Dissecting and refining the staging of chronic kidney disease. Kidney Int 2009; 75: 1009–1014. 2. O’Hare AM, Choi AI, Bertenthal D et al. Age affects outcomes in chronic kidney disease. J Am Soc Nephrol 2007; 18: 2758–2765.
Gianna Mastroianni Kirsztajn1,2, Jose H.R. Suassuna3 and Marcus G. Bastos2,4 1
Division of Nephrology, Department of Medicine, Federal University of Sao Paulo (UNIFESP), Sao Paulo-SP, Brazil; 2Brazilian Society of Nephrology, Committee of Kidney Diseases Prevention/Previna-se Campaign, Sao Paulo-SP, Brazil; 3Department of Internal Medicine, Rio de Janeiro State University (UERJ), Rio de Janeiro-RJ, Brazil and 4Department of Internal Medicine, Federal University of Juiz de Fora (UFJF), Juiz de Fora-MG, Brazil Correspondence: Gianna Mastroianni Kirsztajn, Division of Nephrology, Department of Medicine, Federal University of Sao Paulo, Rua Botucatu 740, Sao Paulo-SP 04023900, Brazil. E-mail: [email protected]
Response to ‘Dividing stage 3 of chronic kidney disease (CKD): 3A and 3B’ Kidney International (2009) 76, 463–464; doi:10.1038/ki.2009.180
We agree that informing unsuspecting patients that they have irreversible ESRD (end-stage renal disease) that requires quick or immediate renal replacement is difficult, and that the task is more awkward if the situation could have been prevented or at least predicted. Kirsztajn et al.1 imply that this, their ‘hardest task,’ can be avoided by the use of the Kidney Disease Outcomes Quality Initiative (KDOQI) staging system. We disagree. In the United Kingdom, in 2007, late presentation of ESRD, also inaccurately called ‘late referral,’ occurred in 21% of the 109 per million population (pmp) new patients starting renal replacement, which translates to B21 pmp Kidney International (2009) 76, 459–464
‘late presenters’ per year.2 If one excludes those patients in whom renal failure and the need for renal replacement could not have been predicted (such as patients with multiple myeloma, vasculitis, catastrophic irreversible acute renal failure, and malignancy), the figure for avoidable late presentation is much lower. Indeed, in 2007, in the Oxford kidney unit, there was a missed opportunity for earlier referral in only 5% of the 150 new patients. The KDOQI staging system would have us believe that 130,000 individuals pmp have chronic kidney disease (CKD). Even if the system could prevent ‘late referral’ (there is no evidence that it does), the labeling of so large a number to identify the 1 in 26,000 CKD patients or 5 per million of the general population who are referred late is not justified. The ‘NNT’ becomes not a ‘number needed to treat,’ but a ‘number needed to terrify.’ In health systems with poor communication between primary and secondary care, the problem of ‘late referral’ may be greater, but it could be solved by educating primary care physicians about identifying those at risk of progressive CKD and how to interpret an abnormal creatinine level. We do not advocate eliminating the early stages of CKD from the system but rather tightening the definition of kidney disease and amalgamating stages 1 and 2, which cannot reliably be distinguished by the Modification of Diet in Renal Disease Study equation for estimated glomerular filtration rate (eGFR).3 Kirsztajn et al. and others also recognize that stage 3 is not homogeneous.4,5 In the United Kingdom, the National Institute for Clinical Excellence suggests subdividing this stage into no less than 4 subcategories, based on whether the eGFR is less or greater than 45 ml/min per m2 and the presence or absence of proteinuria.6 This is a clear admission that the label ‘stage 3 CKD’ is an unsatisfactory and misleading description as pointed out by Kirsztajn et al. Moreover, stage 3 in a 1–5 staging system is halfway to the very end point (ESRD) that they find so painful to discuss and yet, only 0.4% of patients so classified progress to ESRD per annum.7 We are unconvinced by their argument that a change in the staging system would confuse the medical community. Non-nephrologists are already mystified by the so-called epidemic of CKD in their midst, and there are a growing number of calls to correct it.8 If it is wrong, we should admit it and modify it before it becomes embedded in medical thinking. How would medicine ever progress if we accepted theories as truths for fear of causing confusion? We are also unconvinced that doctors would have difficulty in using percentile charts––they do so quite comfortably for assessing growth in children, bone mineral density, and the risks of hyperlipidemia. Indeed, we use the ‘Wetzels’ charts to reassure patients that their eGFRs, although lower than the normal of 90 ml/min per 1.72 m2 asserted by the KDOQI, are in the range for healthy individuals of their age and gender.9 463