Tubular markers do not predict the decline in glomerular filtration rate in type 1 diabetic patients with overt nephropathy

http://www.kidney-international.org & 2011 International Society of Nephrology original article see commentary on page 1042 Tubular markers do not predict the decline in glomerular filtration rate in type 1 diabetic patients with overt nephropathy Stine E. Nielsen 1, Steen Andersen 1, Dietmar Zdunek 2, Georg Hess 3, Hans-Henrik Parving 4,5 and Peter Rossing 1 1 Steno Diabetes Center, Gentofte, Denmark; 2 Roche Diagnostics GmbH, Penzberg, Germany; 3 Department of Internal Medicine, University of Mainz, Mainz, Germany; 4 Department of Medical Endocrinology, Rigshospitalet, University Hospital of Copenhagen, Copenhagen, Denmark and 5 Faculty of Health Sciences, University of Aarhus, Aarhus, Denmark Recent studies have shown that both glomerular and tubulointerstitial damage are important factors in the pathophysiology and progression of diabetic nephropathy. To examine whether markers of tubular damage are useful in monitoring the progression of disease, we measured urinary levels of neutrophil gelatinase-associated lipocalin (NGAL), liver fatty acid-binding protein (LFABP), and kidney injury molecule-1 (KIM-1) in a 3-year intervention study of 63 type 1 diabetic patients with kidney disease. The baseline mean glomerular filtration rate (GFR) was 87 ml/min per 1.73 m 2 and urinary albumin excretion 1141 mg/24 h. Patients with the highest compared with the lowest quartile of urinary NGAL at baseline had higher urinary KIM-1 levels and a significant decrease in their GFR each year. Using linear regression analysis, we found that elevated urinary NGAL and KIM-1 concentrations were associated with a faster decline in GFR, but not after adjustment for known promoters of progression. Urinary LFABP was not related to decline in GFR. Losartan treatment (100 mg/day) reduced urinary KIM-1 by 43% over a 12-month period. Thus, urine biomarker measurements in patients with type 1 diabetic nephropathy did not provide additional prognostic information to that of known progression promoters. Kidney International (2011) 79, 1113 1118; doi:10.1038/ki.2010.554; published online 26 January 2011 KEYWORDS: diabetes mellitus; diabetic nephropathy; tubule cells Correspondence: Stine E. Nielsen, Steno Diabetes Center, Niels Steensens Vej 2, Gentofte 2820, Denmark. E-mail: Sene@steno.dk Received 12 May 2010; revised 8 November 2010; accepted 7 December 2010; published online 26 January 2011 Diabetic nephropathy is a serious complication with high morbidity and mortality. One-third of patients with diabetic nephropathy develop end-stage renal disease (ESRD), and the cumulative incidence of ESRD is 8% at 30 years after onset of type 1 diabetes. 1 As an attempt to prevent progression from diabetic nephropathy to ESRD, numerous progression promoters have been investigated and found to be essential targets for intervention. This includes albuminuria, arterial blood pressure, blood lipids, restriction of protein intake, and glycemic control. 2 Despite improved outcome due to aggressive multifactorial intervention, many patients still develop ESRD. 3 Recently, studies have shown that not only glomerular but also tubulointerstitial damage are important factors in the pathophysiology and progression of diabetic nephropathy. 4,5 Markers of tubular damage could, therefore, potentially be useful in the evaluation of prognosis and for monitoring the effect of intervention. Different markers of tubular damage have been suggested in recent years: Some of the first and most prominent were a1- and b2-microglobulin, and since then many markers of tubular damage has been discovered. Neutrophil gelatinase-associated lipocalin (NGAL) is a small protein that belongs to the lipocalin protein family. NGAL is produced in epithelia cells and neutrophils in most tissues and is a marker of renal tubular damage. 6 Mori and Nakao 7 presented a review on the forest fire theory : They see the ongoing nephron damage as buring trees, whereas trees/ nephrons which are already burned down/atrophic is not producing NGAL, so NGAL is a result of the active damage: the nephrons that can still be saved. They argue that NGAL is representative of the functioning tubular mass and produced as a response to tubular injury. Liver fatty acid-binding protein (LFABP) is an intracellular carrier protein also produced in proximal tubular cells, as a response to tubulointerstitial damage, and it reflects accurately the degree of tubular injury. 8 Another marker is kidney injury molecule- 1 (KIM1), which has been shown to be a marker of tubular damage in various chronic kidney diseases. 9 Kidney International (2011) 79, 1113 1118 1113

original article SE Nielsen et al.: Tubular markers and progression of diabetic nephropathy Previously, we have shown that levels of the tubular markers, urinary (u)-ngal 10 and u-lfabp, 11,12 are increased in type 1 diabetic patients, even before they develop signs of glomerular damage, that is, micro- or macroalbuminuria. However, the role of u-ngal and u-lfabp in the further progression from diabetic nephropathy toward ESRD in type 1 diabetes is unknown. In the current work, we aimed to determine the role of u-ngal, u-kim1, and u-lfabp as predictors of decline in glomerular filtration rate (GFR) in patients with type 1 diabetes and diabetic nephropathy. This was carried out in a 3-year prospective intervention trial, initially designed to investigate the long-term renoprotective effects of losartan in 63 patients with type 1 diabetes and diabetic nephropathy, with different angiotensin-converting enzyme ID genotypes, 13,14 In previous studies it has been shown that renin angiotensin aldosteron system blockade reduces apoptotic factors in tubules and thereby preserves tubular function. 15 On the basis of this hypothesis we investigated the effect of the angiotensin II receptor blocker losartan on the tubular markers. RESULTS Sixty-three patients completed the study. Twenty-five patients were women. Mean (s.d.) age was 44 (9) years and mean duration of diabetes was 32 (9) years. At baseline, GFR was 87 (24) ml/min per 1.73 m 2 and the level of albuminuria (geometric mean (95% confidence interval)): 1141 (930 1401) mg/24 h. Wash out (4 weeks) 0 Losartan 50 mg (2 months) Baseline Losartan100 mg and additional antihypertensive treatment (permitted after 4 months) 1 2 3 U-KIM1, u-ngal and u-lfabp, GFR, and 24 hr blood pressure Figure 1 Design of the study. Years At baseline, after the wash out period (Figure 1), u-ngal was 1878 (1268 2781) (ng)/(g), u-lfabp was 1300 (743 2270) (ng/g), and u-kim1 was 59 (48 72) (ng/g). U-KIM1 and u-ngal at baseline were associated with u-albumin at baseline (all log transformed; R ¼ 0.44, P ¼ 0.002 and R ¼ 0.31, P ¼ 0.03), respectively: whereas u-lfabp was not associated with u-albumin (R ¼ 0.02). Baseline levels of the three tubular markers were not associated with baseline GFR (P40.4). Decline in GFR Decline in GFR during the 3-year study period was 4.1 (3.6) ml/min per 1.73 m 2 per year. We found that patients with u- NGAL in the highest versus the lowest quartile at baseline had a decline in GFR of 6.8 versus 2.6 ml/min per 1.73 m 2 (P ¼ 0.01), and for u-kim1 5.5 versus 2.6 (P ¼ 0.04; see Table 1). In a linear regression analyses, high levels of u-ngal (R ¼ 0.32, P ¼ 0.02) and u-kim1 (R ¼ 0.24, P ¼ 0.1) at baseline were associated with a faster decline in GFR, but the association were not significant when adjusted for known progression promoters as seen in Table 2 (sex, age, HbA 1c, systolic blood pressure, urinary albumin excretion rate). See Figures 2 and 3 for scatter plots. U-LFABP did not predict decline in GFR (R ¼ 0.1, P ¼ 0.5). Effect of losartan From baseline visit, without losartan treatment to full dose 100 mg at 1 year, the effects of losartan were evaluated (Figure 1 and Table 3). As previously reported, losartan significantly reduced blood pressure and albuminuria. For this study, we found that u-kim1 was significantly reduced by 43 (29 52)% and stayed at this level during 3-year followup (see Figure 4a and b). The decline in u-kim1 from 0 to 12 months was associated with change in u-albumin (R ¼ 0.51, Po0.001), but not with systolic blood pressure (R ¼ 0.16, P ¼ 0.6), or decline in GFR during the study (R ¼ 0.14, P ¼ 0.4). U-NGAL and u-lfabp were not significantly affected by losartan treatment (increased 1% ( 39 to 29%) and 34% ( 34 to 170%), respectively). The levels of u-lfabp and u-ngal did not change significantly during the 3-year follow-up. DISCUSSION We found that in type 1 diabetic patients with diabetic nephropathy high levels of the markers of tubular damage Table 1 Yearly decline in GFR according to baseline levels of markers of tubular damage GFR decline (mean, s.d.)/year ml/min per 1.73 m 2 First quartile Second quartile Third quartile Fourth quartile P-value ANOVA/(lower vs upper quartile) u-ngal* 2.6 (1.9) 4.0 (3.7) 3.9 (3.3) 6.8 (4.4) 0.037/0.01 u-kim1* 2.6 (2.8) 4.1 (2.9) 5.2 (4.7) 5.5 (3.7) 0.20/0.04 u-lfabp* 4.5 (3.4) 3.3 (3.9) 4.6 (4.7) 5.2 (3.1) 0.30/0.69 Abbreviations: ANOVA, analysis of variance; GFR, glomerular filtration rate; u-kim1, urinary kidney injury molecule-1; u-lfabp, urinary liver fatty acid-binding protein; u-ngal, urinary neutrophil gelatinase-associated lipocalin. Analysis performed by ANOVA followed by comparison of the difference of decline in GFR between the upper and lower quartile. *ng/g. 1114 Kidney International (2011) 79, 1113 1118

SE Nielsen et al.: Tubular markers and progression of diabetic nephropathy original article Table 2 Multiple regression analyses illustrating the influence of log-transformed u-ngal, u-kim1, and u-alb and known confounders on GFR decline per year U-NGAL U-KIM1 U-Alb Multiple R 0.33 0.33 0.54 0.24 0.25 0.54 0.43 b (s.e.) b (s.e.) b (s.e.) b (s.e.) b (s.e.) b (s.e.) b (s.e.) Tubular marker 2.0* (0.87) 2.1* (0.9) 0.46 (1.2) 2.9 (1.7) 2.8 (1.8) 0.12 (2.3) Sex 0.15 (0.2) 1.1 (1.4) 0.4 (1.2) 1.3 (1.3) Age 0.01 (0.06) 0.01 (0.07) 0.01 (0.06) 0.01 (0.8) Diabetes duration 0.05 (0.07) 0.1 (0.07) HbA1c 0.35 (0.5) 0.34 (0.5) Log u-albumin 4.0 (1.9) 4.1 (2.0) 4.4* (1.2) Systolic blood pressure 0.05 (0.06) 0.05 (0.06) Abbreviations: GFR, glomerular filtration rate; u-alb, urinary albumin; u-kim1, urinary kidney injury molecule-1; u-ngal, urinary neutrophil gelatinase-associated lipocalin. Reported values are b (-unstandardized) (s.e.), *Po0.05. Decline GFR/year (ml/min/m 2 ) 15.00 10.00 5.00 0.00 Decline in GFR/year (ml/min/m 2 ) 15.00 10.00 5.00 0.00 2.00 0.00 Log10 (u-ngal) Figure 2 Scatter plot of baseline urinary neutrophil gelatinase-associated lipocalin (u-ngal) versus decline in glomerular filtration rate (GFR) during the study period (R ¼ 0.32, P ¼ 0.02). 2.00 0.0 0.5 1.0 Log10 (u-kim1) Figure 3 Scatter plot of baseline urinary kidney injury molecule-1 (u-kim1) versus decline in glomerular filtration rate (GFR) during the study period (R ¼ 0.24, P ¼ 0.1). 1.5 u-ngal and u-kim1, but not u-lfabp, were predictive of faster decline in GFR during 3-year of follow-up after onset of antihypertensive treatment with the angiotensin II receptor blocker losartan. However, the markers did not provide additional prognostic information, as the association between baseline levels of the markers and decline in GFR was not independent of known progression promoters. U-KIM1 was significantly reduced after onset of losartan treatment, and this reduction was associated with the decline in albuminuria. In a study of 103 diabetic patients with varying degrees of glomerulosclerosis, it was reported that estimated GFR was not always related to glomerular damage. When there were no tubulointerstitial damage, the patients had preserved kidney function despite the glomerular damage, whereas a reduced kidney function was observed in cases with tubulointerstitial damage. 16 This indicates that tubular damage is not just a passive result of glomerular damage, but has an independent role in the development of renal impairment. However, this statement has been questioned and another hypothesis suggested that the glomerular damage and tubular damage are involved in early and late damage in diabetic nephropathy, respectively. 4 Mori and Nakao 7 suggested that u-ngal is a real time indicator of renal damage: the forest fire theory as described in the introduction. They propose that u-ngal do not, as other markers of kidney damage, reflect the number of remaining nephrons, but instead is a dynamic measure of activity of disease. This conclusion was based on animal studies. In 96 patients with chronic kidney disease, it has been shown by Bolignano et al. 17 that u-ngal is predictive of progression of chronic kidney disease (doubling of s-creatinine/esrd) even after adjustment for other progression promoters (adjusted for serum NGAL, estimated GFR, and age). However, in their study only 20% of the patients had diabetic nephropathy, the other 80% had non-diabetic nephropathy of various etiology. Our study extends their findings as we also observed an association between u-ngal and decline in GFR; however, in our study this was not independent of other progression promotors. Their follow- Kidney International (2011) 79, 1113 1118 1115

original article SE Nielsen et al.: Tubular markers and progression of diabetic nephropathy Table 3 Effects of losartan on laboratory values measured at baseline and after 12 months treatment in 63 type 1 diabetic patients with diabetic nephropathy Losartan 0 mg (baseline) Losartan 100 mg (1 year) P-value Systolic blood pressure (mm Hg) 152 (17) 142 (16) o0.001 Diastolic blood pressure (mm Hg) 80 (9) 73 (8) o0.001 GFR (ml/min per 1.73 m 2 ) 87 (24) 78 (23) o0.001 u-albumin (mg/24 h) a 1141 (930 1401) 469 (348 632) o0.001 u-kim1 a (ng/g) 59 (48 72) 38 (31 47) o0.001 u-lfabp a (ng/g) 1300 (743 2270) 1484 (884 249) 0.42 u-ngal a (ng/g) 1878 (1268 2781) 1794 (1305 2466) 0.95 Abbreviations: GFR, glomerular filtration rate; u-kim1, urinary kidney injury molecule-1; u-lfabp, urinary liver fatty acid-binding protein; u-ngal, urinary neutrophil gelatinase-associated lipocalin. a Geometric mean (95% confidence interval). Mean log10 (u-kim1) 0.8 0.6 0.4 0.2 0.0 U-KIM1 reduction (%) from baseline Baseline 0 10 20 30 40 50 60 12 Months 24 Months 36 Months 12 Months 24 Months 36 Months Figure 4 Effect of losartan on tubular markers. (a) Reductions in log-transformed urinary kidney injury molecule-1 (u-kim1) during losartan treatment. KIM1 is significantly decreased from baseline (no difference from 1 to 3 years). Error bars indicate 95% confidence interval. (b) Reductions in log-transformed u-kim1 during losartan treatment, expressed in percentages. up time was only 18 months compared with 3 years of follow-up in this study. More important is that they reported estimated GFR compared with our true GFR measured with plasma clearance of an external marker (51Cr-EDTA). NGAL has mainly been investigated in late chronic kidney disease, as recently reviewed by Bolignano et al., 18 and not much in early nephropathy. In cross-sectional studies of patients with earlier stages of diabetic nephropathy, we and others have found that u-lfabp is associated with u-albumin excretion levels (R ¼ 0.73, Po0.001) and thereby reflects the degree of kidney damage. 12 In a prospective study of type 1 diabetic normoalbuminuric patients followed for a median of 18 years, we found that u-lfabp predicted development of diabetic nephropathy. 11 These patients were free of renal complications at baseline. As u-lfabp levels were not predictive of decline in GFR in the present study of patients with overt diabetic nephropathy, this supports the idea that the tubular phase or damage reflected by u-lfabp is more important in the early stages of diabetic nephropathy, and thus the value of this tubular marker is greatest at that stage. In this study we found, in a linear regression analysis, that u-ngal and u-kim1 at baseline predicted the rate of decline in GFR during 3 years of follow-up. This seems to be mediated through the known progression promoters, as the difference disappears after adjustment for these variables. This suggests that the tubular damage reflected by increased levels of these markers is important in the development and further progression to ESRD. However, the fact that they are not independently able to predict decline in GFR indicates that there is no additional information obtained by measuring them in this situation. The major change in the model was seen when entering u-albumin excretion rate, as the P-value for u-ngal was changed to 0.7, indicating that the effect of u-ngal does not add to the effect of u-albumin (see Table 3). This is in accordance with the correlation between u-ngal and urinary albumin excretion, although the correlation was not strong. We have not been able to find any studies, besides the present, in overt diabetic nephropathy describing the relationship between u-ngal/u- LFABP/u-KIM1 and kidney function, as expressed by urinary albumin excretion or GFR. Urinary excretion of KIM1 is not only known to be closely related to both the tubular production of KIM1 but is also specific to tubular damage, as confirmed by biopsies in both clinical and experimental studies. 9,19 Another interesting finding is that in studies of human biopsies with tubular necrosis or atrophic and fibrotic damage, KIM1 is not produced. The authors, therefore, conclude that KIM1 is a good marker of active/ongoing tubular damage and not tubular scarring. 9 We found that the angiotensin II receptor 1116 Kidney International (2011) 79, 1113 1118

SE Nielsen et al.: Tubular markers and progression of diabetic nephropathy original article blocker losartan reduces u-kim1 significantly with 43%. Experimental studies in diabetic rats have shown that renin angiotensin aldosteron system blockade reduces apoptotic factors and thereby preserves tubular function. 15 It has been hypothesized that the reduced tubular damage seen when blocking the renin angiotensin system is due to reduced oxidative stress and correction of chronic hypoxia, independently of the blood pressure-lowering effect. 15 This is in accordance with findings in non-diabetic nephropathy. In a study of 34 non-diabetic patients with proteinuria, losartan was found to reduce u-kim1 excretion with B30%, 20 this was highly associated with the decrease in proteinuria. The authors speculated that the treatment induced reduction in albuminuria, which led to reduced albumin toxicity on the tubules. However in this study, only part of the reduction in u-kim1 can be explained by reduced tubular albumin toxicity (R ¼ 0.53, Po0.001), and furthermore the reduction was not related to decline in GFR. In our study, we hypothesized that patients with the largest decrease in tubular markers would benefit most (smallest decline in GFR) during follow-up time. However, this was not the case (change in u-kim1 versus GFR decline during the study: R ¼ 0.14, P ¼ 0.4). Baseline u-kim1 (that is, tubular damage) itself was a better marker than change, which could not only reflect tubular repair (and our study could lack the power to show such an effect), but also changes induced by losartan, which were not related to effect on decline in GFR, in which case an association between change and decline could not be expected. Finally, it is not possible to exclude that the association with baseline KIM1 could be due to confounding. U-LFABP and u-ngal were not reduced during longterm losartan treatment in this study. We have previously found that in type 1 diabetic patients with nephropathy, treatment with the angiotensin-converting enzyme inhibitor lisinopril for 8 weeks reduced u-lfabp by 40% (Po0.05), 12 independently of the reduction in urine albumin excretion, and u-ngal with 17% (not significant). 10 Whether the difference between the studies could be explained by the use of angiotensin-converting enzyme inhibitor versus angiotensin II receptor blocker, or rather if there is a short-term effect of renin angiotensin aldosteron system blockade on the markers, which disappears with time, will need to be investigated further in future studies. To our knowledge this is the first prospective study in diabetic nephropathy comparing u-ngal, u-kim1, and u-lfabp with true GFR measurements based on plasma clearance of 51 Cr EDTA and a long follow-up period. Other studies in chronic kidney disease have primarily been crosssectional, and prospective studies have mainly been performed in non-diabetic patients with chronic kidney disease with change in estimated GFR as the end point. It is a limitation that we did not have access to renal biopsies in the subjects, as the diagnosis was made clinically according to established criteria. If biopsies had been available, they could have potentially confirmed our hypothesis that the renal production of the tubular markers are closely related to the urinary excretion and that the tubular markers reflect the tubular damage and tubular functioning mass accurately. Moreover, this could have resulted in evaluation of the functioning glomerular mass. This would be interesting and relevant to look into in future studies. However, as mentioned previously, this has been demonstrated in several previous studies. 8,9,18 A potential limitation of this study is the storage of the samples at 20 1C for 10 years before analysis; however, all samples were stored in similar conditions. It has previously been shown that u-ngal is stable after long-term storage at 20 1C. 21 Regarding u-lfabp, it has been shown to be stable after freezing (personal communication T Sugaya, unpublished data). Another limitation is that we only had one urine per visit available for tubulus marker measurement. In conclusion, we found that type 1 diabetic patients with high levels of u-ngal and u-kim1 are associated with a faster decline in GFR, suggesting that tubular damage is important for progression, even though the markers did not add information about progression or treatment effect beyond known progression promoters. Furthermore, patients treated with the angiotensin II receptor blocker, losartan, had a significant reduction in u-kim1. More studies are needed to determine the role of tubular damage, expressed by excretion of u-ngal, u-kim1, and u-lfabp, in diabetic nephropathy. MATERIALS AND METHODS The study was a 3-year prospective intervention trial initially designed to investigate the long-term renoprotective effects of losartan in 63 patients with type 1 diabetes and diabetic nephropathy, with different angiotensin-converting enzyme ID genotypes. 13 For this study, we determined u-lfabp, u-ngal, and u-kim1 at baseline and yearly during follow-up (for design, see Figure 1). Diabetic nephropathy was diagnosed clinically in patients with persistent albuminuria (4300 mg/24 h), diabetic retinopathy, and absence of other evidence of kidney or renal tract disease. Before enrollment, all antihypertensive medication was withdrawn for at least 4 weeks. All patients fulfilled the following inclusion criteria: diabetic nephropathy, GFR 460 ml/min per 1.73 m 2, office blood pressure 4135/85 mm Hg, and age between 18 and 70 years. Dietary intake of protein or salt was not restricted. All antihypertensive medication was withdrawn for at least 4 weeks before enrolment. Thereafter, at baseline, patients were treated with losartan at a dose of 50 mg once daily for the first 2 months and 100 mg once daily for the remainder of the study. Additional antihypertensive treatments, that is, diuretics, calcium channel blockers, and a-blockers, were given in an attempt to achieve a target blood pressure of below 135/85 mm Hg. Blood pressure measurements and adjustments of antihypertensive treatment were determined every 3 months. Glomerular filtration rate, u-albumin excretion rate, and 24 h blood pressure were determined at baseline and every 6 months thereafter, as described previously. 13 Laboratory variables were determined at baseline as well as at the first (after 2 months) and last visit during losartan treatment. Kidney International (2011) 79, 1113 1118 1117

original article SE Nielsen et al.: Tubular markers and progression of diabetic nephropathy Glomerular filtration rate was measured as plasma clearance of an intravenous injection of 51 Cr-EDTA. 22 The results were standardized for 1.73 m 2 body surface area, using the same surface for each patient during the study. The mean coefficient of variation in GFR of each patient from day to day was 4%. Albumin excretion rate was determined as the geometric mean of at least two consecutive 24-h urine collections, completed at baseline and immediately before each visit. The tubular markers, u-ngal, u-lfabp, and u-kim1 (ELISA, Roche, Penzberg, Germany), were measured in one 24 h urine sample at baseline, after wash out and yearly thereafter and expressed relative to u-creatinine. From venous blood samples, HbA 1c, serum potassium, sodium, creatinine, and cholesterol concentrations were determined, and blood pressure values were determined based on 24-h ambulatory blood pressure measurements performed using the Takeda TM2420 device (version 7; A&D, Tokyo, Japan; for details see ref. 13). This study was performed according to the principles of the Declaration of Helsinki and approved by the ethical committee of Copenhagen County. All patients gave their informed consent. Statistic analysis Data are expressed as means (s.d.), except for u-albumin, u-lfabp, u-ngal, and u-kim1, which were logarithmically transformed owing to their skewed distribution, and given as the geometric means (95% confidence interval). The rate of decline in kidney function was analyzed with regression lines for GFR over the followup period using all measurements of GFR during the study period. Comparisons of normally or log-normally distributed variables were performed using Student s t-test. When quartiles were compared, analysis of variance with subsequent comparison of bottom versus top quartile was performed. Univariate and multivariate linear regression analysis were used to determine the association between tubular markers and decline in GFR. Association between changes in tubular markers and changes in urinary albumin excretion and GFR were evaluated with linear regression. We adjusted for the following known progression promotors in our analysis: sex, age, diabetes duration, HbA 1c, systolic blood pressure, and urinary albumin excretion rate. A P-value o0.05 was considered significant (two tailed). Data were analyzed using the commercially available statistical program SPSS 14.0 software program (SPSS, Chicago, IL). DISCLOSURE DZ is current employee and GH is previous employee of Roche Diagnositics. GH is a consultant to Roche. All the other authors declared no competing interests. ACKNOWLEDGMENTS We thank TR Juhl, BR Jensen, UM Smidt, and I-L Rossing for laboratory assistance. REFERENCES 1. Finne P, Reunanen A, Stenman S et al. Incidence of end-stage renal disease in patients with type 1 diabetes. JAMA 2005; 294: 1782 1787. 2. Parving H-H, Jacobsen P, Rossing K et al. Benefits of long-term antihypertensive treatment on prognosis in diabetic nephropathy. Kidney Int 1996; 49: 1778 1782. 3. Gaede P, Lund-Andersen H, Parving H-H et al. Effect of a multifactorial intervention on mortality in type 2 diabetes. N Engl J Med 2008; 358: 580 591. 4. Parving H-H, Mauer M, Ritz E. Diabetic nephropathy. In: Brenner BM (ed). Brenner & Rector s THE KIDNEY, 8th edn. WB Saunders: Boston, USA, 2008, pp 1265 1298. 5. Gilbert RE, Cooper ME. The tubulointerstitium in progressive diabetic kidney disease: More than an aftermath of glomerular injury? Kidney Int 1999; 56: 1627 1637. 6. Cowland JB, Sorensen OE, Sehested M et al. Neutrophil gelatinaseassociated lipocalin is up-regulated in human epithelial cells by IL-1 beta, but not by TNF-alpha. J Immunol 2003; 171: 6630 6639. 7. Mori K, Nakao K. Neutrophil gelatinase-associated lipocalin as the real-time indicator of active kidney damage. Kidney Int 2007; 71: 967 970. 8. Kamijo A, Sugaya T, Hikawa A et al. Urinary excretion of fatty acid-binding protein reflects stress overload on the proximal tubules. Am J Pathol 2004; 165: 1243 1255. 9. van Timmeren MM, van den Heuvel MC, Bailly V et al. Tubular kidney injury molecule-1 (KIM-1) in human renal disease. J Pathol 2007; 212: 209 217. 10. Nielsen SE, Schjoedt KJ, Astrup AS et al. Neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule 1 (KIM1) in patients with diabetic nephropathy: a cross-sectional study and the effects of lisinopril. Diabet Med 2010; 27: 1144 1150. 11. Nielsen SE, Sugaya T, Hovind P et al. Urinary liver-type fatty acid-binding protein (u-lfabp) predicts progression to nephropathy in type 1 diabetic patients. Diabetes Care 2010; 33: 1320 1324. 12. Nielsen SE, Sugaya T, Tarnow L et al. Tubular and glomerular injury in diabetes and the impact of ACE inhibition. Diabetes Care 2009; 32: 1684 1688. 13. Andersen S, Tarnow L, Cambien F et al. Long-Term renoprotective effects of losartan in diabetic nephropathy: interaction with ACE insertion/ deletion genotype? Diabetes Care 2003; 26: 1501 1506. 14. Schjoedt KJ, Andersen S, Rossing P et al. Aldosterone escape during blockade of the renin-angiotensin-aldosterone system in diabetic nephropathy is associated with enhanced decline in glomerular filtration rate. Diabetologia 2004; 47: 1936 1939. 15. Kelly DJ, Cox AJ, Tolcos M et al. Attenuation of tubular apoptosis by blockade of the renin-angiotensin system in diabetic Ren-2 rats. Kidney Int 2002; 61: 31 39. 16. Bader R, Bader H, Grund KE et al. Structure and function of the kidney in diabetic glomerulosclerosis: correlations between morphological and functional parameters. Pathol Res Pract 1980; 167: 204 216. 17. Bolignano D, Lacquaniti A, Coppolino G et al. Neutrophil gelatinaseassociated lipocalin (NGAL) and progression of chronic kidney disease. Clin J Am Soc Nephrol 2009; 4: 337 344. 18. Bolignano D, Donato V, Coppolino G et al. Neutrophil gelatinaseassociated lipocalin (NGAL) as a marker of kidney damage. Am J Kidney Dis 2008; 52: 595 605. 19. Vaidya VS, Ramirez V, Ichimura T et al. Urinary kidney injury molecule-1: a sensitive quantitative biomarker for early detection of kidney tubular injury. Am J Physiol Renal Physiol 2006; 290: F517 F529. 20. Waanders F, Vaidya VS, van GH et al. Effect of renin-angiotensinaldosterone system inhibition, dietary sodium restriction, and/or diuretics on urinary kidney injury molecule 1 excretion in nondiabetic proteinuric kidney disease: a post hoc analysis of a randomized controlled trial. Am J Kidney Dis 2009; 53: 16 25. 21. Wagener G, Lee HT. In Reply to [ ]instability of urinary NGAL during longterm storage and [ ]NGAL and cardiac surgery-associated acute kidney injury. Am J Kidney Dis 2009; 53: 566. 22. Bröchner-Mortensen J, Rödbro P. Selection of routine method for determination of glomerular filtration rate in adult patients. Scand J Clin Lab Invest 1976; 36: 35 45. 1118 Kidney International (2011) 79, 1113 1118