Fenofibrate reduces progression to microalbuminuria over 3 years in a placebo-controlled study in type 2 diabetes: Results from the Diabetes Atherosclerosis Intervention Study (DAIS)

Fenofibrate Reduces Progression to Microalbuminuria Over 3 Years in a Placebo-Controlled Study in Type 2 Diabetes: Results From the Diabetes Atherosclerosis Intervention Study (DAIS) Jean-Claude Ansquer, MD, Christelle Foucher, PhD, Stephanie Rattier, MS, Marja-Riitta Taskinen, MD, and George Steiner, MD, for the DAIS Investigators ● Background: Microalbuminuria is an early marker of diabetic nephropathy and an independent risk factor for cardiovascular disease. In the Diabetes Atherosclerosis Intervention Study (DAIS), treatment of people with type 2 diabetes with micronized fenofibrate for an average of 38 months reduced the progression of angiographically evaluated coronary artery disease and improved lipoprotein level abnormalities compared with placebo. The aim of this analysis is to study the influence of the treatment on changes in urinary albumin excretion. Methods: Microalbuminuria was measured on 2 to 3 occasions by using timed overnight samples at baseline and yearly thereafter in 314 DAIS participants (77 women, 237 men; average age, 56 years); all except 3 participants had either a normal albumin excretion rate (<20 ␮g/min; n ⴝ 214) or microalbuminuria (albumin, 20 to 200 ␮g/min; n ⴝ 97) before randomization. Tabulated shifts (between normal, microalbuminuria, and macroalbuminuria) from baseline to last observed values were compared between treatment groups by means of chi-square or Fisher’s exact test. Results: Fenofibrate significantly reduced the worsening of albumin excretion (fenofibrate, 8% versus placebo, 18%; P < 0.05). This effect was caused mostly by reduced progression from normal albumin excretion to microalbuminuria: 3 of 101 participants in the fenofibrate group versus 20 of 113 participants in the placebo group (P < 0.001). Overall, changes in albumin excretion were independent of age or changes in lipid or creatinine levels, weight, or blood pressure. Conclusion: Improvement in lipid profiles with fenofibrate in patients with type 2 diabetes was associated with reduced progression from normal albumin excretion to microalbuminuria. Am J Kidney Dis 45:485-493. © 2005 by the National Kidney Foundation, Inc. INDEX WORDS: Albumin excretion; type 2 diabetes; fenofibrate.

M

ICROALBUMINURIA is a predicator of both nephropathy and cardiovascular disease in people with and without diabetes mellitus.1-3 It is crucial to search for interventions to reduce both the development and progression of diabetic renal disease and associated cardiovascular risk. Control of glycemia and blood pressure are conventional interventions in patients with diabetes that reduce the risk for neuropathy, retinopathy, and nephropathy.4 Conversely, improvement of nephropathy through lipid control, as hypothesized by Moorehead et al,5 remains unproven. The Steno-2 study of participants with type 2 diabetes and microalbuminuria showed that intensive treatment with a stepwise implementation of behavior modifications and pharmacological therapies, including lipid-lowering treatments, reduced the risk for cardiovascular and macrovascular events by approximately 50% compared with conventional therapy.6 Limited information is available on the natural history and effect of pharmacological intervention in people with diabetes and normal albumin excretion. In 176 participants with type 2 diabetes and normoalbuminuria, the 5-year cumulative incidence of the development of microalbumin-

uria (albumin, 30 to 299 mg/24 h) was 23%, and each 39-mg/dL (1-mmol/L) increase in serum cholesterol concentration was associated with a significant 40% increase in risk for the development of microalbuminuria.7 The Diabetes Atherosclerosis Intervention Study (DAIS) showed that correcting

From the Departments of Clinical Research and Medical Affairs, and Biometrics, Fournier Pharma, Daix, France; Biomedicum Helsinki, Helsinki University Central Hospital, Helsinki, Finland; and Toronto General Hospital, Toronto, Ontario, Canada. Received July 22, 2004; accepted in revised form November 6, 2004. Originally published online as doi:10.1053/j.ajkd.2004.11.004 on January 14, 2005. A full list of DAIS investigators is provided in Lancet 357:905-910, 2001. J-C.A., C.F., and S.R. are employees of Laboratoires Fournier SA, which manufactures fenofibrate. M-R.T. and G.S. are part of the DAIS study group that received unrestricted grants from Laboratoires Fournier SA to conduct the trial. Address reprint requests to Jean-Claude Ansquer, MD, Department of Clinical Research and Medical Affairs, Laboratoires Fournier SA, 50 Rue de Dijon, 21121 Daix, France. E-mail [email protected] © 2005 by the National Kidney Foundation, Inc. 0272-6386/05/4503-0005$30.00/0 doi:10.1053/j.ajkd.2004.11.004

American Journal of Kidney Diseases, Vol 45, No 3 (March), 2005: pp 485-493

485

486

ANSQUER ET AL Table 1. Clinical Characteristics of Participants With Normal Albumin Excretion, Microalbuminuria, and Macroalbuminuria All Participants

Male/female (%) Age (y) BMI (kg/m2) SBP/DBP (mm Hg) Previous coronary intervention (%) Current smokers (%) Ex-smokers (%) Duration of diabetes (y)

Normal Albumin Excretion (n ⫽ 214)

Microalbuminuria and Macroalbuminuria (n ⫽ 100)

Fenofibrate (n ⫽ 155)

Placebo (n ⫽ 159)

72/28 57 ⫾ 6 28.7 ⫾ 3.1 137 ⫾ 17/80 ⫾ 8 31 14 53 8.4 ⫾ 6.2

82/18 56 ⫾ 6 29.6 ⫾ 2.8* 146 ⫾ 19/85 ⫾ 9† 36 23 47 9.0 ⫾ 5.6

74/26 57 ⫾ 6 29.0 ⫾ 3.1 139 ⫾ 19/82 ⫾ 9 34 15 50 8.8 ⫾ 6.2

77/23 56 ⫾ 6 29.1 ⫾ 3.1 141 ⫾ 18/81 ⫾ 9 31 19 52 8.5 ⫾ 5.7

NOTE. Values expressed as mean ⫾ SD or percent. Normal albumin excretion, urinary albumin less than 20 ␮g/min; microalbuminuria and macroalbuminuria, urinary albumin of 20 ␮g/min or greater. Previous coronary interventions are percutaneous transluminal coronary angioplasty and coronary artery bypass graft. SBP/DBP missing data: n ⫽ 4 for normal albumin excretion, n ⫽ 2 for the fenofibrate and placebo groups. *P ⬍ 0.05 versus normal albumin excretion. †P ⬍ 0.001 versus normal albumin excretion.

for lipid level abnormalities in participants with type 2 diabetes after fenofibrate treatment for at least 3 years was associated with a significant reduction in the progression of focal coronary atherosclerosis disease.8 This study also provided the opportunity to determine whether correction of lipid level abnormalities by means of fenofibrate treatment also reduced the progression of proteinuria in this population, mostly composed of normoalbuminuric participants. METHODS

Fournier SA, Daix, France) treatments for a minimum 3-year duration. Details of blood sampling and laboratory analytic and standardization procedures also have been described.10 Of 418 participants randomized in DAIS, 314 participants had urinary albumin measurements available to compute shifts between normal albumin excretion, microalbuminuria, and macroalbuminuria from baseline and last observed value during treatment period (study end). Most participants (86%) had the last value observed at 3 years or later. Participants were divided according to baseline albumin excretion rate: normoalbuminuria (albumin ⬍ 20 ␮g/min), microalbuminuria (albumin, 20 to 200 ␮g/min), and macroalbuminuria (albumin ⱖ 200 ␮g/min).

Study Design and Population

Statistical Analyses

This study was performed in a representative sample of DAIS participants. DAIS is a randomized, double-blind, placebo-controlled, angiographic study. Participants were selected to be men or women of middle age, to have undergone coronary intervention or not, and to have mild to moderate lipid level abnormalities typical of patients with type 2 diabetes. Selection criteria and baseline characteristics have been published.9 Among selection criteria, participants were not to have significant renal disease, shown by a history of proteinuria with protein of 500 mg/24 h or greater or albumin excretion of 200 ␮g or greater and/or creatinine levels of 1.7 mg/dL or greater (ⱖ150 ␮mol/L) in men and 1.6 mg/dL or greater (ⱖ140 ␮mol/L) in women. Fasting blood samples were collected and albumin excretion was determined in overnight timed urine collections performed by the participant following written and oral instructions. Collections were performed on 2 or 3 occasions during the baseline period (8, 4, and 2 weeks before randomization) and at yearly intervals after randomization to placebo or 200 mg of micronized fenofibrate (Laboratoires

Continuous data are provided as mean ⫾ SD and geometric mean for data with skewed distribution. Baseline values are the mean of all available measurements during the run-in period. Measured variables were compared between groups by using Student’s t-test or Mann-Whitney-Wilcoxon test. Categorical variables were compared by using chi-square or Fisher’s exact test. Categorical shifts between normal albumin excretion, microalbuminuria, and macroalbuminuria from baseline and study end were compared according to treatment groups by using chi-square or Fisher’s exact test. Distribution of participants showing progression, regression, or no change in albumin excretion also was compared between treatment groups by using chi-square or Fisher’s exact test without and with controlling for either baseline blood pressure, smoking status, or glycemic control by means of the Cochran-Mantel-Haenszel test. Univariate correlations between albumin excretion and demographic, anthropometric, or lipid measurements were assessed by using Pearson’s parametric correlation coefficient. Statistical analyses were performed using SAS, release 8.2, software (SAS Institute, Cary, NC).

FENOFIBRATE SLOWS PROGRESSION TO MICROALBUMINURIA

487

Table 2. Biochemical Characteristics of Participants With Normal Albumin Excretion, Microalbuminuria, and Macroalbuminuria All Participants

HbA1c (%) Fasting glucose (mg/dL) TG (mg/dL) Total cholesterol (mg/dL) Low-density lipoprotein cholesterol (mg/dL) HDL-C (mg/dL) Men Women Creatinine (mg/dL) Men Women Urinary albumin (␮g/min) Total homocysteine (mg/L) Fibrinogen (g/L)

Normal Albumin Excretion (n ⫽ 214)

Microalbuminuria and Macroalbuminuria (n ⫽ 100)

Fenofibrate (n ⫽ 155)

Placebo (n ⫽ 159)

7.4 ⫾ 1.2 157 ⫾ 47 195 214 ⫾ 29 131 ⫾ 28 40 ⫾ 8 39 ⫾ 7 42 ⫾ 8 0.9 ⫾ 0.2 1.0 ⫾ 0.1 0.8 ⫾ 0.1 7.3 1.37 3.2 ⫾ 0.7

8.0 ⫾ 1.1* 178 ⫾ 43* 232† 216 ⫾ 29 127 ⫾ 29 38 ⫾ 8‡ 37 ⫾ 8‡ 43 ⫾ 8 1.0 ⫾ 0.2 1.0 ⫾ 0.2 0.8 ⫾ 0.1 49.7 1.45 3.4 ⫾ 0.7‡

7.6 ⫾ 1.1 163 ⫾ 45 211 215 ⫾ 30 130 ⫾ 28 39 ⫾ 7 38 ⫾ 7 42 ⫾ 8 1.0 ⫾ 0.2 1.0 ⫾ 0.2 0.8 ⫾ 0.1 14.3 1.42 3.3 ⫾ 0.8

7.6 ⫾ 1.3 168 ⫾ 49 202 214 ⫾ 28 129 ⫾ 28 40 ⫾ 8 39 ⫾ 8 43 ⫾ 8 0.9 ⫾ 0.1 1.0 ⫾ 0.1 0.8 ⫾ 0.1 12.7 1.37 3.3 ⫾ 0.7

NOTE. Values expressed as mean ⫾ SD except for TG, urinary albumin, and total homocysteine, which are expressed as geometric mean. Normal albumin excretion, urinary albumin less than 20 ␮g/min; microalbuminuria and macroalbuminuria, urinary albumin of 20 ␮g/min or greater. Total homocysteine missing data, n ⫽ 19 and n ⫽ 12 for normal and abnormal albumin excretion, respectively; n ⫽ 14 and n ⫽ 17 for the fenofibrate and placebo groups, respectively; fibrinogen missing data, n ⫽ 7 for normal albumin excretion, n ⫽ 4 and n ⫽ 3 for the fenofibrate and placebo groups, respectively. HbA1c and fasting glucose missing data, n ⫽ 1 for normal albumin excretion and the placebo group. To convert glucose in mg/dL to mmol/L, multiply by 0.05551; TG in mg/dL to mmol/L, multiply by 0.01129; total cholesterol, low-density lipoprotein cholesterol, and HDL-C in mg/dL to mmol/L, multiply by 0.02586; creatinine in mg/dL to ␮mol/L, multiply by 88.4; total homocysteine in mg/L to ␮mol/L, multiply by 7.397; fibrinogen in g/L to mg/dL, multiply by 100. *P ⬍ 0.001 versus normal albumin excretion. †P ⬍ 0.01 versus normal albumin excretion. ‡P ⬍ 0.05 versus normal albumin excretion.

RESULTS

Baseline Characteristics of Participants There were no differences in clinical and biochemical characteristics between the placebo and fenofibrate groups (Tables 1 and 2). There was no significant difference in age between participants with normal and abnormal albumin excretion or in the distribution of men, participants with previous coronary intervention, and smoking status (Table 1). However, body mass index (BMI) and blood pressure were significantly greater in the group with abnormal albumin excretion than in the group with normal albumin excretion (P ⬍ 0.05 and P ⬍ 0.001, respectively; Table 1). Levels of fasting triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C), fasting glucose, and hemoglobin A1c (HbA1c) also were significantly different between participants with normal and abnormal albumin excretion (Table 2). At baseline, 24% of participants

(25% and 21% in the placebo and fenofibrate groups, respectively) were administered an angiotensin-converting enzyme (ACE) inhibitor (72 patients) or angiotensin II receptor blocker (ARB; 1 patient). Albumin Excretion at Baseline and Its Evolution During the Course of the Study One hundred thirteen participants (71%) had normal albumin excretion at baseline in the placebo group versus 101 participants (65%) in the fenofibrate group (Table 3). Forty-four (28%) and 53 participants (34%) were reported to have microalbuminuria at baseline in the placebo and fenofibrate groups, respectively. Three participants (2 participants, placebo group; 1 participant, fenofibrate group) were reported to have macroalbuminuria. After treatment, 107 (67%) and 118 participants (76%) were reported to have normal albumin excretion in the placebo and

488

ANSQUER ET AL Table 3. Shift in Albumin Excretion From Baseline to Study End in the Placebo and Fenofibrate Groups Study End

Placebo No. of patients Baseline Normal (n ⫽ 113) Microalbuminuria (n ⫽ 44) Macroalbuminuria (n ⫽ 2) Fenofibrate No. of patients Baseline Normal (n ⫽ 101) Microalbuminuria (n ⫽ 53) Macroalbuminuria (n ⫽ 1)

Normal

Microalbuminuria

Macroalbuminuria

107

44

8

92 15 0

20 22 2

1 7 0

118

27

10

98 20 0

3 24 0

0 9 1

NOTE. Values are number of participants. Microalbuminuria, urinary albumin between 20 and 200 ␮g/min; macroalbuminuria, urinary albumin of 200 ␮g/min or greater.

fenofibrate groups; 44 (28%) and 27 participants (17%) with microalbuminuria and 8 (5%) and 10 participants (7%) with macroalbuminuria, respectively (Table 3). Progression of albumin excretion is defined as a shift between normal albumin excretion to either microalbuminuria or macroalbuminuria or from microalbuminuria to macroalbuminuria. Regression is defined as a shift between microalbuminuria or macroalbuminuria to normal albumin excretion or macroalbuminuria to microalbuminuria. There was a significant difference (P ⫽ 0.031) in the 3 categories between treatment groups (Fig 1). There were fewer participants with progression in albumin excretion in the fenofibrate than placebo group: 12 (8%) versus 28 participants (18%). After adjustment for blood pressure, the difference in distribu-

tion of progressors, regressors, and unchanged participants remained significant between treatment groups (P ⫽ 0.037). A similar result was observed after adjustment for either smoking status or baseline HbA1c level or baseline blood glucose level. Three and 20 participants with normoalbuminuria at baseline were reported at study end with microalbuminuria in the fenofibrate and placebo groups, respectively (P ⬍ 0.001). Characteristics of Participants Showing Progression, Regression, or No Change in Albumin Excretion The evolution of BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), and levels of HbA1c, TG, urinary albumin, creatinine, total homocysteine, and fibrinogen in particiFig 1. Percentage of participants with progression, regression, or no change in albumin excretion during the course of the study. Progression is a shift between normal albumin excretion to either microalbuminuria or macroalbuminuria or between microalbuminuria to macroalbuminuria from baseline to study end. Regression is a shift between microalbuminuria or macroalbuminuria to normal albumin excretion or macroalbuminuria to microalbuminuria from baseline to study end.

FENOFIBRATE SLOWS PROGRESSION TO MICROALBUMINURIA

pants showing progression (ie, progressors), regression (ie, regressors), or no change (ie, unchanged) in urinary albumin excretion after placebo and fenofibrate treatments are listed in Table 4. After fenofibrate treatment, participants classified as progressors had a greater BMI and SBP than those classified as regressors and unchanged. TG levels were decreased in all categories, with the greatest reduction observed in regressors (Table 4). The greatest increases in urinary albumin and creatinine levels were observed in participants with the greatest level of these parameters at baseline, ie, in those classified as progressors (Table 4). In these participants, urinary albumin level at baseline increased to macroalbuminuria after either placebo or fenofibrate treatment (Table 4). Regarding the distribution of ACE inhibitor and ARB treatments, 32% of participants (35% and 28% in the placebo and fenofibrate groups, respectively) were treated at study end. Most participants were administered ACE inhibitors. Only 10 and 4 participants in the placebo and fenofibrate groups were administered ARBs, respectively. Univariate Correlations At baseline, albumin excretion significantly correlated positively with weight (r ⫽ 0.17; P ⬍ 0.05), BMI (r ⫽ 0.18; P ⬍ 0.01), SBP and DBP (r ⫽ 0.29; P ⬍ 0.001 for SBP; r ⫽ 0.23; P ⬍ 0.001 for DBP), and HbA1c (r ⫽ 0.16; P ⬍ 0.01), TG (r ⫽ 0.13; P ⬍ 0.05), creatinine (r ⫽ 0.12; P ⬍ 0.05), total homocysteine (r ⫽ 0.16; P ⬍ 0.01), and fibrinogen levels (r ⫽ 0.12; P ⬍ 0.05). No correlation was observed between urinary albumin level and duration of diabetes (r ⫽ 0.07; P ⫽ 0.242). At the last visit in the placebo group, albumin excretion still correlated significantly with SBP and DBP (r ⫽ 0.17; P ⬍ 0.05; r ⫽ 0.20; P ⬍ 0.05, respectively) and HbA1c level (r ⫽ 0.19; P ⬍ 0.05). At the last visit in the fenofibrate group, albumin excretion correlated significantly with creatinine level (r ⫽ 0.26; P ⬍ 0.001), SBP (r ⫽ 0.18; P ⬍ 0.05), and BMI (r ⫽ 0.17; P ⬍ 0.05). However, in both groups, changes in urinary albumin levels were not related to changes in creatinine, homocysteine, and fibrinogen levels. Changes in urinary albumin levels correlated significantly with changes in urinary creatinine levels (r ⫽ 0.19; P ⬍ 0.05) in the placebo group and changes in weight (r ⫽

489

0.16; P ⬍ 0.05) in the fenofibrate group. In this group, changes in urinary albumin also correlated significantly with changes in HDL-C levels (r ⫽ 0.18; P ⬍ 0.05), but not with changes in TG levels (r ⫽ ⫺0.09; P ⫽ 0.276). DISCUSSION

The present study shows that treatment of people with type 2 diabetes with micronized fenofibrate reduces the progression of urinary albumin excretion. This was evidenced by a reduction in number of participants with microalbuminuria at the end of the treatment period and less progression from normoalbuminuria to microalbuminuria (Table 3) in a representative subset of participants with clinical and biochemical characteristics close to those of the DAIS population.8 In the placebo group, the rate of progression from normal albumin excretion during 3 years compared well with the figure of 23% during 5 years reported by Gall et al7 in a diabetic population of similar age, BMI, and blood pressure. Participants who had macroalbuminuria at study end or who progressed to macroalbuminuria did not differ between the 2 treatment groups. Although significant correlations between urinary albumin levels and weight, BMI, blood pressure, and HbA1c, TG, and creatinine levels were observed at baseline, no significant correlation between changes in urinary albumin levels and changes in any of these parameters except for weight changes in the fenofibrate group was observed after treatment in the present study. A low proportion of participants administered ACE inhibitors or ARBs was observed at baseline; therefore, initiation of antihypertensive therapy during the treatment period may have altered the progression of albumin excretion. However, use of these treatments was not significantly different between the fenofibrate and placebo groups at study end. Furthermore, after controlling for baseline blood pressure, treatment effect remained statistically significant, suggesting that the observed reduction in progression to microalbuminuria induced by fenofibrate treatment was independent of hypertensive status. Smoking status also is a strong risk factor in the progression of albuminuria. Controlling for smoking status still led to a significant effect of fenofibrate treatment in reducing the progression

490

ANSQUER ET AL

Table 4. Evolution of BMI, Blood Pressure, HbA1c, TG, Urinary Albumin, Creatinine, Total Homocysteine, and Fibrinogen Levels in Participants With Progression, Regression, or No Change in Albumin Excretion After Placebo and Fenofibrate Treatment Progressors

Placebo No. of patients BMI (kg/m2) Baseline Study end SBP (mm Hg) Baseline Study end DBP (mm Hg) Baseline Study end HbA1c (%) Baseline Study end TG (mg/dL) Baseline Study end Urinary albumin (␮g/min) Baseline Study end Creatinine (mg/dL) Baseline Study end Total homocysteine (mg/L) Baseline Study end Fibrinogen (g/L) Baseline Study end Fenofibrate No. of patients BMI (kg/m2) Baseline Study end SBP (mm Hg) Baseline Study end DBP (mm Hg) Baseline Study end HbA1c (%) Baseline Study end TG (mg/dL) Baseline Study end Urinary albumin (␮g/min) Baseline Study end Creatinine (mg/dL) Baseline Study end Total homocysteine (mg/L) Baseline Study end

Regressors

28

Unchanged

Total

17

114

29.5 ⫾ 3.1 29.5 ⫾ 3.4

29.8 ⫾ 3.0 29.7 ⫾ 3.6

29.0 ⫾ 3.1 29.1 ⫾ 3.3

147 ⫾ 20 144 ⫾ 21

142 ⫾ 18 138 ⫾ 18

139 ⫾ 17 141 ⫾ 16.5

84 ⫾ 8 80 ⫾ 13

84 ⫾ 6 79 ⫾ 8

80 ⫾ 9 81 ⫾ 9

81.4 ⫾ 8.5 80.7 ⫾ 9.4

8.1 ⫾ 1.1 8.5 ⫾ 1.6

7.8 ⫾ 1.2 8.0 ⫾ 1.6

7.5 ⫾ 1.3 7.6 ⫾ 1.2

7.6 ⫾ 1.3 7.8 ⫾ 1.4

196 161

217 212

201 172

159 29.1 ⫾ 3.1 29.2 ⫾ 3.3 140.8 ⫾ 17.6 141.3 ⫾ 17.5

202 174

19.3 75.9

39.8 13.9

9.6 10.4

12.7 15.2

1.0 ⫾ 0.2 1.0 ⫾ 0.2

1.0 ⫾ 0.1 1.0 ⫾ 0.3

0.9 ⫾ 0.1 0.9 ⫾ 0.2

0.9 ⫾ 0.1 0.9 ⫾ 0.2

1.41 1.50 3.3 ⫾ 0.7 3.6 ⫾ 0.7

1.49 1.65 3.3 ⫾ 0.6 3.2 ⫾ 0.7

12

1.35 1.35 3.3 ⫾ 0.7 3.2 ⫾ 0.7

1.37 1.41 3.3 ⫾ 0.7 3.3 ⫾ 0.7

20

123

29.7 ⫾ 3.4 30.2 ⫾ 3.5

28.7 ⫾ 2.3 28.0 ⫾ 2.2

28.9 ⫾ 3.2 29.1 ⫾ 3.4

29.0 ⫾ 3.1 29.0 ⫾ 3.3

144 ⫾ 20 146 ⫾ 23

146 ⫾ 20 145 ⫾ 17

137 ⫾ 19 141 ⫾ 19

139.2 ⫾ 19.4 141.7 ⫾ 18.9

84 ⫾ 9 78 ⫾ 11

86 ⫾ 10 79 ⫾ 8

81 ⫾ 9 80 ⫾ 8

81.8 ⫾ 9.1 79.6 ⫾ 8.4

7.6 ⫾ 1.2 8.0 ⫾ 1.5

7.7 ⫾ 1.0 7.9 ⫾ 1.0

7.6 ⫾ 1.1 8.0 ⫾ 1.3

7.6 ⫾ 1.1 8.0 ⫾ 1.3

260 155

233 111

50.8 222.1

203 136

155

211 134

35.0 8.9

10.9 10.1

14.3 12.7

1.1 ⫾ 0.2 1.4 ⫾ 0.4

0.9 ⫾ 0.14 1.1 ⫾ 0.3

1.0 ⫾ 0.18 1.1 ⫾ 0.3

1.0 ⫾ 0.2 1.1 ⫾ 0.3

1.65 2.46

1.28 2.00

(Continued)

1.42 2.05

1.42 2.08

FENOFIBRATE SLOWS PROGRESSION TO MICROALBUMINURIA

491

Table 4 (Cont’d). Evolution of BMI, Blood Pressure, and HbA1c, TG, Urinary Albumin, Creatinine, Total Homocysteine, and Fibrinogen Levels in Participants With Progression, Regression, or No Change in Albumin Excretion After Placebo and Fenofibrate Treatment

Fibrinogen (g/L) Baseline Study end

Progressors

Regressors

Unchanged

Total

3.5 ⫾ 0.6 3.3 ⫾ 0.7

3.2 ⫾ 0.6 2.9 ⫾ 0.4

3.3 ⫾ 0.8 2.9 ⫾ 0.6

3.3 ⫾ 0.8 2.9 ⫾ 0.6

NOTE: Values expressed as mean ⫾ SD, except for TG, urinary albumin, and total homocysteine, which are expressed as geometric mean. Progression is a shift between normal albumin excretion to either microalbuminuria or macroalbuminuria or between microalbuminuria to macroalbuminuria from baseline to study end. Regression is a shift between microalbuminuria or macroalbuminuria to normal albumin excretion or macroalbuminuria to microalbuminuria from baseline to study end. Missing data for the placebo group: SBP, n ⫽ 2 in progressors; HbA1c, n ⫽ 1 in regressors; homocysteine, n ⫽ 4, n ⫽ 2, and n ⫽ 11 in progressors, regressors, and unchanged, respectively; fibrinogen, n ⫽ 1 and n ⫽ 2 in progressors and unchanged, respectively. Missing data for the fenofibrate group: SBP, n ⫽ 2 in progressors; homocysteine: n ⫽ 1, n ⫽ 1, and n ⫽ 12 in progressors, regressors, and unchanged, respectively; fibrinogen, n ⫽ 4 in unchanged. To convert TG in mg/dL to mmol/L, multiply by 0.01129; creatinine in mg/dL to ␮mol/L, multiply by 88.4; total homocysteine in mg/L to ␮mol/L, multiply by 7.397; fibrinogen in g/L to mg/dL, multiply by 100.

of urinary albumin excretion. Similar results were observed after adjustment for either baseline HbA1c or blood glucose level, suggesting that the effect of fenofibrate treatment was independent of baseline glycemic control. As anticipated in the placebo group, participants with greater HbA1c levels and elevated blood pressure at baseline were more likely to progress. This was corroborated by the significant correlation between final urinary albumin level and either HbA1c level or SBP and DBP. A previous meta-analysis suggested that the use of lipid-lowering agents could reduce the rate of progression of renal disease.11 However, to date, this suggestion remains speculative because of the strong heterogeneity in data caused, in part, by different methods used to measure proteinuria between studies. Nevertheless, recent analyses of the Cholesterol and Recurrent Events study showed an inverse relation between moderate chronic renal insufficiency (glomerular filtration rate ⬍ 60 mL/min/1.73 m2 body surface area) before treatment and slowing of renal function loss with pravastatin treatment, with a lesser reduction in glomerular filtration rate with pravastatin compared with placebo in patients with a glomerular filtration rate less than 40 mL/min and those with proteinuria at baseline.12 Conversely, the Prevention of Renal and Vascular Endstage Disease Intervention Trial failed to show a benefit of pravastatin in 854 participants with microalbuminuria, whereas fosinopril, 20 mg, was effective in this 4-year factorial design study.13

One limitation in that study is the definition of macroalbuminuria, ie, albumin excretion from 15 to 300 mg/24 h, which included participants considered normoalbuminuric because mean urinary albumin excretion at baseline ranged from 22 to 24 mg/24 h. Lipid and lipoprotein level abnormalities, in particular, elevated intermediate-density lipoprotein-cholesterol level, reduced HDL-C levels, and elevated remnant-like particle levels, were observed in participants with diabetes with various degrees of nephropathy14 or persistent microalbuminuria15 and overt proteinuria.16 Thus, corrections of these lipoprotein level abnormalities by means of lipid-lowering therapy may contribute to the prevention of more severe nephropathy. Deighan et al16 observed a significant reduction in remnant-like particle cholesterol (⫺35%) and remnant-like particle TG levels (⫺44%) in participants with proteinuria in the nephrotic range after fenofibrate treatment for 2 months. However, they did not evaluate the relation between these changes and changes in protein excretion after treatment. In the present study, a significant correlation was observed between urinary albumin and TG levels at baseline; TG levels were greater in participants with abnormal urinary albumin excretion compared with those with normoalbuminuria. After fenofibrate treatment, changes in urinary albumin excretion correlated positively with changes in HDL-C level, but not with changes in TG level, which averaged a 30% reduction in the DAIS study. Nevertheless, participants

492

showing regression of albumin excretion tended to have lower TG levels at study end than those progressing (Table 4). In a small placebocontrolled study of Japanese hypertriglyceridemic participants, a trend to a reduction in urinary albumin excretion was reported after 2 months of fenofibrate treatment in association with a reduction in serum TG, apolipoprotein CII, and apolipoprotein CIII levels.17 A similar trend also was observed previously in participants with type 2 diabetes treated for 1 year with gemfibrozil.18 Improvement in lipid profiles and endothelial function after a 3-month treatment with fenofibrate has been shown in normoalbuminuric participants with type 2 diabetes19 and subjects with hypertriglyceridemia.20 Because endothelial dysfunction appears to predate the development of microalbuminuria,21 a possible explanation for the observed reduction in progression to microalbuminuria after fenofibrate treatment could be an improvement in vascular function. However, this has not been studied to date. Potential limitations of the study should be considered. Blood pressure and glucose level were adequately, but not optimally, controlled during the study. Moreover, the lack of urinary albumin measurements after the study did not permit the evaluation of whether a reduction in progression to microalbuminuria was persistent after withdrawal of fenofibrate treatment. Finally, the number of participants, their characteristics typical of an early stage of type 2 diabetes, and duration of study end were too small to assess an effect on similar end points as those assessed in ARB trials. In conclusion, the present study shows that fenofibrate treatment for at least 3 years is effective in reducing the progression of renal disease in people with type 2 diabetes without diabetic nephropathy. The explanation for such an effect of a lipidlowering drug, independent of changes in blood pressure, smoking status, or glycemic control, remains unclear and thus needs to be investigated further, as well as its consequences on renal function, in larger clinical trials with long-term fenofibrate treatment. The possibility of a causal effect of the nonlipid pleiotropic effects of fenofibrate also should be considered.

ANSQUER ET AL

REFERENCES 1. Mogensen CE: Microalbuminuria predicts clinical proteinuria and early mortality in maturity-onset diabetes. N Engl J Med 310:356-360, 1984 2. Beilin J, Stanton KG, McCann VJ, Knuiman MW, Divitini ML: Microalbuminuria in type 2 diabetes: An independent predictor of cardiovascular mortality. Aust N Z J Med 26:519-525, 1996 3. Diercks GF, van Boven AJ, Hillege JL, de Jong PE, Rouleau JL, van Gilst WH: The importance of microalbuminuria as a cardiovascular risk indicator: A review. Can J Cardiol 18:525-535, 2002 4. Hollenberg NK: Treatment of the participant with diabetes mellitus and risk of nephropathy. What do we know, and what do we need to learn? Arch Intern Med 164:125130, 2004 5. Moorehead JF, Chan MK, El-Nahas M, Varghese Z: Lipid nephrotoxicity in chronic progressive glomerular and tubulo-interstitial disease. Lancet 11:1309-1311, 1982 6. Gaede P, Vedel P, Larsen N, Jensen GVH, Parving H-H, Pedersen O: Multifactorial intervention and cardiovascular disease in participants with type 2 diabetes. N Engl J Med 348:383-393, 2003 7. Gall MA, Hougaard P, Borch-Johnsen K, Parving H-H: Risk factors for development of incipent and overt diabetic nephropathy in participants with non-insulin dependent diabetes mellitus: Prospective, observational study. BMJ 314: 783-788, 1997 8. Diabetes Atherosclerosis Intervention Study Investigators: Effect of fenofibrate on progression of coronary artery disease in type 2 diabetes: The Diabetes Atherosclerosis Intervention Study. Lancet 357:905-910, 2001 9. Steiner G: The Diabetes Atherosclerosis Intervention study (DAIS): A study conducted in cooperation with the World Health Organization. The DAIS Project Group. Diabetologia 39:1655-1661, 1996 10. McGuinness C, Seccombe DW, Frohlich JJ, Ehnholm C, Sundval J, Steiner G: Laboratory standardisation of a large international clinical trial: The DAIS experience. DAIS Project Group. Diabetes Atherosclerosis Intervention Study. Clin Biochem 33:15-24, 2000 11. Fried LF, Orchard TJ, Kasiske BL: Effect of lipid reduction on the progression of renal disease: A metaanalysis. Kidney Int 59:260-269, 2001 12. Tonelli M, Moye L, Sacks FM, Cole T, Curhan GC, Cholesterol and Recurrent Events Trial Investigators: Effect of pravastatin on loss of renal function in people with moderate chronic renal insufficiency and cardiovascular disease. J Am Soc Nephrol 14:1605-1613, 2003 13. Asselbergs FW, Diercks GFH, Hillege HL, et al for the Prevention of Renal and Vascular Endstage Disease Intervention Trial (PREVEND IT) Investigators: Effects of Fosinopril and Pravastatin on cardiovascular events in subjects with macroalbuminuria. Circulation 110:2809-2816, 2004 14. Shoji T, Emoto M, Kawagishi T, et al: Atherogenic lipoprotein changes in diabetic nephropathy. Atherosclerosis 156:425-433, 2001

FENOFIBRATE SLOWS PROGRESSION TO MICROALBUMINURIA

15. Watts GF, Naumova R, Slavin BM, et al: Serum lipids and lipoproteins in insulin-dependent diabetic participants with persistent microalbuminuria. Diabet Med 6:25-30, 1989 16. Deighan CJ, Caslake M, McConnell M, BoultonJones JM, Packard CJ: Comparative effects of cerivastatin and fenofibrate on the atherogenic lipoprotein phenotype in proteinuric renal disease. J Am Soc Nephrol 12:341-348, 2001 17. Kasumi T, Hirano T, Yoshino G, for the Fenofibrate Study Group: Effects of fenofibrate on albuminuria in participants with hypertriglyceridemia and/or hyperuricemia: A multicenter, randomised, double-blind, placebo-controlled, crossover study. Curr Ther Res Clin Exp 64:434-446, 2003 18. Smulders YM, van Eeden AE, Stehouwer CD, Weijers RN, Slaats EH, Silberbusch J: Can reduction in hypertriglyceri-

493

daemia slow progression of microalbuminuria in patients with non-insulin-dependent diabetes mellitus? Eur J Clin Invest 27:997-1002, 1997 19. Playford DA, Watts GF: Fenofibrate improves endothelial function in dyslipidaemic non-insulin dependent diabetes (NIDDM). Atherosclerosis 151:S209A, 2000 (suppl 1; abstr) 20. Capell WH, DeSouza CA, Poirier P, et al: Short-term triglyceride lowering with fenofibrate improves vasodilator function in subjects with hypertriglyceridemia. Arterioscler Thromb Vasc Biol 23:307-313, 2003 21. Meeking DR, Cummings MH, Thorne S, et al: Endothelial dysfunction in type 2 diabetic subjects with and without microalbuminuria. Diabet Med 16:841-847, 1999