Clinical and Experimental Nephrology

, Volume 15, Issue 3, pp 391–397

Risk of macrovascular disease stratified by stage of chronic kidney disease in type 2 diabetic patients: critical level of the estimated glomerular filtration rate and the significance of hyperuricemia

Authors

    • Division of Diabetes and MetabolismThe Institute for Adult Diseases, Asahi Life Foundation
    • Department of NephrologyToho University School of Medicine
  • Shigeko Hara
    • Division of Diabetes and MetabolismThe Institute for Adult Diseases, Asahi Life Foundation
    • Kidney Center and Okinaka Memorial Institute for Medical ResearchToranomon Hospital
  • Akifumi Kushiyama
    • Division of Diabetes and MetabolismThe Institute for Adult Diseases, Asahi Life Foundation
  • Yoshifumi Ubara
    • Kidney Center and Okinaka Memorial Institute for Medical ResearchToranomon Hospital
  • Yoko Yoshida
    • Division of Diabetes and MetabolismThe Institute for Adult Diseases, Asahi Life Foundation
  • Sonoo Mizuiri
    • Department of NephrologyToho University School of Medicine
  • Atsushi Aikawa
    • Department of NephrologyToho University School of Medicine
  • Shouji Kawatzu
    • Division of Diabetes and MetabolismThe Institute for Adult Diseases, Asahi Life Foundation
Original Article

DOI: 10.1007/s10157-011-0420-6

Cite this article as:
Tanaka, K., Hara, S., Kushiyama, A. et al. Clin Exp Nephrol (2011) 15: 391. doi:10.1007/s10157-011-0420-6

Abstract

Background

Although a high prevalence of macrovascular disease (MVD) has been reported in patients with stage 3 chronic kidney disease (CKD), few studies have reported its risk with respect to the underlying cause of kidney disease. This study investigated the prevalence of MVD in type 2 diabetic patients with CKD stratified by CKD stage, as defined by estimated glomerular filtration rate (eGFR), as well as the risk factors for MVD.

Methods

1493 patients with diabetic CKD (1273 males, 220 females) were stratified by CKD stage (stage 1: 39, stage 2: 272, stage 3: 1052, stage 4: 101, stage 5: 29) based on eGFR calculated by the Japanese formula and averaged over 8 months. MVD was defined as one of the following: coronary heart disease (CHD), stroke or arteriosclerosis obliterans (ASO).

Results

The prevalence of MVD was 18.6%. A significant increasing trend in MVD prevalence was observed from stage 3 (17.78%) to 4 (52.48%). According to a receiver operating characteristic curve analysis on MVD prevalence in stage 3 patients, an eGFR of 46.4 ml/min/1.73 m2 was determined to be a critical cut-off level. Proteinuria, eGFR <60 ml/min/1.73 m2 and hyperuricemia were independent risk factors for MVD.

Conclusions

In patients with diabetic CKD, a significant increase in MVD prevalence was observed from stage 3 to 4. An eGFR of 46.4 ml/min/1.73 m2 is a critical level that affects MVD prevalence. From the perspective of cardiorenal association, CKD stage 3 should be divided into two substages. As hyperuricemia is related to an increased risk of MVD, uric acid control may be important in reducing MVD risk in diabetic CKD.

Keywords

Type 2 diabetesChronic kidney diseaseMacrovascular diseaseHyperuricemia

Introduction

Chronic kidney disease (CKD) [1] is diagnosed from renal function tests and the evidence of renal injuries, such as urinary abnormal findings. In the definition of CKD, the cause of kidney disease is not taken into account. Diabetic nephropathy is the most common cause of CKD, so this population is important from the perspective of cardiorenal association [24] and prevention of end-stage renal disease (ESRD). In general, the rate of cardiovascular disease increases after CKD stage 2, and cardiovascular death is higher than the rate of the progression to ESRD [4, 5]; however, few studies have examined cardiovascular disease in diabetic CKD patients stratified by CKD stage, and the clinical characteristics and risk factors for each stage have not been clarified. This retrospective study examined the clinical findings and prevalence of macrovascular disease (MVD) in diabetic CKD patients and investigated the relationship between estimated glomerular filtration rate (eGFR) and MVD prevalence. Moreover, risk factors for MVD were analyzed in a cross-sectional study.

Subjects and methods

Subjects

Among 1950 patients with type 2 diabetes who were followed at the Institute for Adult Diseases, Asahi Life Foundation, between January 1 and August 31, 2008, 1493 patients with CKD (mean age 64.9 ± 10.1 years; disease duration 18.5 ± 9.5 years; 1273 males and 220 females) were studied. The patients were stratified by CKD stage and clinical characteristics. Age, diabetes duration, body mass index (BMI), blood pressure, smoking history and laboratory findings (HbA1c, HDL cholesterol, LDL cholesterol, triglycerides, serum uric acid, creatinine, eGFR and hemoglobin) were observed. The presence of concurrent vascular diseases, including diabetic retinopathy, coronary heart disease (CHD), stroke, and arteriosclerosis obliterans (ASO), were investigated and compared among different CKD stages. The relationship between MVD prevalence and eGFR was analyzed and risk factors were examined. BMI was calculated by dividing weight in kilograms by the square of height in meters. The eGFR used for CKD staging was calculated by the estimation formula advocated by the Japanese Society of Nephrology, as follows [6]:
$$ {\text{eGFR}}\left( {{\text{ml}}/\min/1.73\;{\text{m}}^{2} } \right) = 194 \times {\text{Cr}}^{ - 1.094} \times {\text{age}}^{ - 0.287} \left( { \times 0.739\;{\text{if}}\;{\text{female}}} \right). $$

Definitions of diseases

CKD was defined as eGFR <60 ml/min/1.73 m2 and/or microalbuminuria or overt proteinuria in our study.

Diabetic retinopathy was diagnosed with a history of retinal bleeding. Hypertension was defined as systolic blood pressure ≥130 mmHg, and/or diastolic blood pressure ≥80 mmHg, and/or current use of antihypertensive medication. Hyperlipidemia was defined as LDL cholesterol level ≥120 mg/dl and/or current use of antihyperlipidemic medication. Hyperuricemia was defined as serum uric acid level ≥7.0 mg/dl for men or 6.0 mg/dl for women and/or current use of antihyperuricemic medication. CHD was defined as a diagnosis of significant stenosis by cardiac catheterization and/or a history of catheter intervention. Stroke was defined as a history of cerebral bleeding and/or cerebral infarction, including lacunar infarction confirmed by computed tomography (CT) or magnetic resonance imaging (MRI). ASO was defined as ankle brachial pressure index (ABI) <0.9 and/or a diagnosis by angiography, including enhanced CT and MRI, and/or a history of catheter or surgical treatment. MVD was defined as having at least one of the three vascular diseases: CHD, stroke or ASO.

Statistical analyses

The Cochran–Armitage test for trend and the Ryan method were used to compare clinical findings among CKD stages. The eGFR that predicts MVD prevalence in each stage was delineated using a receiver operating characteristic (ROC) analysis. Risk factors of MVD were identified using a multivariate analysis. The software JMP (version 8; SAS Institute, Cary, USA) was used for statistical analyses. A p value less than 0.05 was considered significant. The mean values of the laboratory data between January and August 2008 were used.

Results

The clinical characteristics of 1493 patients with type 2 diabetic CKD are shown in Table 1. The mean age was 64.9 years, and the mean diabetes duration was 18.5 years. Systolic blood pressure (mean ± SD) was 130 ± 14 mmHg, diastolic blood pressure was 73 ± 10 mmHg, HbA1c was 7.2 ± 1.1%, and eGFR was 51.3 ± 16.5 ml/min/1.73 m2. The prevalence of hypertension was 65.1%, hyperlipidemia 59%, hyperuricemia 23.7%, retinopathy 49.9%, CHD 10.5%, stroke 9.0%, and ASO 3.9%. The prevalence of MVD was 18.6%.
Table 1

Clinical characteristics of type 2 diabetic CKD patients

Age (year)

64.9 ± 10.1

Retinopathy (%)

49.9% (n = 746)

Diabetes duration (year)

18.5 ± 9.5

Hypertension (%)

65.1% (n = 972)

Body mass index (kg/m2)

24.2 ± 3.8

Hyperlipidemia (%)

59.0% (n = 881)

Systolic BP (mmHg)

130 ± 14

Hyperuricemia (%)

23.7% (n = 354)

Diastolic BP (mmHg)

73 ± 10

Normoalbuminuria (%)

36.7% (n = 548)

Hemoglobin A1C (%)

7.2 ± 1.1

Microalbuminuria (%)

40.2% (n = 600)

HDL-cholesterol (mg/dl)

53.2 ± 13.7

Macroalbuminuria (%)

23.1% (n = 345)

LDL-cholesterol (mg/dl)

106.9 ± 24.6

CHD (%)

10.5% (n = 157)

Triglyceride (mg/dl)

168.3 ± 107.6

Stroke (%)

9.0% (n = 134)

Uric acid (mg/dl)

5.5 ± 1.2

ASO (%)

3.9% (n = 58)

Creatinine (mg/dl)

0.94 ± 0.48

MVD (%)

18.6% (n = 279)

eGFR (ml/min/1.73 m2)

51.3 ± 16.5

  

Hemoglobin (g/dl)

13.8 ± 1.5 (n = 1388)

  

Hypertension was defined as systolic blood pressure ≥130 mmHg, and/or diastolic blood pressure ≥80 mmHg, and/or current use of antihypertensive medication

Hyperlipidemia was defined as LDL-C level ≥120 mg/dl and/or current use of antihyperlipidemic medication

Hyperuricemia was defined as serum uric acid level ≥7.0 mg/dl in male, 6.0 mg/dl in female, and/or current use of antihyperuricemic medication

CHD was defined as a previous history of myocardial infarction or angina pectoris, confirmed by coronary interventions

Stroke was defined as a previous history of bleeding or ischemic stroke included lacuna infarction, confirmed by brain CT or MRI

ASO was diagnosed by angiography including enhanced CT or MRI and/or ankle-brachial pressure index (ABI) <0.9

MVD was defined as having one of the vascular diseases (CHD or stroke or ASO)

CHD coronary heart disease, ASO arteriosclerosis obliterans, MVD macrovascular disease

The clinical characteristics of the patients stratified by CKD stage into stages 1 to 5 were analyzed and compared. As the CKD stage progressed, increases in age, diabetes duration, serum uric acid, creatinine and anemia, as well as decreases in blood pressure, BMI, and HbA1c were observed (Table 2). As CKD stage progressed, the prevalence of smoking history, hyperuricemia and retinopathy were significant increased (Cochran–Armitage test for trend: p < 0.0001). A definitive trend was not observed for the prevalence of hypertension and hyperlipidemia.
Table 2

Patients’ characteristics stratified by CKD stages

CKD stage (n = 1493)

1 (n = 39)

2 (n = 272)

3 (n = 1052)

4 (n = 101)

5 (n = 29)

Trend p

Male:female (1273:220)

8:31

185:187

966:86

88:13

26:3

 

Age (year)

59.4 ± 12.5

61.8 ± 11.6

65.2 ± 9.5

69.9 ± 8.7

69.3 ± 8.0

<0.0001

Diabetes duration (year)

16.6 ± 10.3

16.4 ± 8.1

18.4 ± 9.5

24.0 ± 10.0

24.9 ± 9.0

<0.0001

Body mass index (kg/m2)

25.9 ± 9.6

24.6 ± 3.8

24.0 ± 3.4

24.7 ± 3.4

23.1 ± 3.0

0.001

Systolic BP (mmHg)

136 ± 14

132 ± 12

128 ± 14

132 ± 18

133 ± 15

<0.0001

Diastolic BP (mmHg)

76 ± 9

76 ± 9

73 ± 9

70 ± 11

67 ± 11

<0.0001

Hemoglobin A1C (%)

8.2 ± 1.6

7.5 ± 1.4

7.1 ± 1.0

7.0 ± 1.0

6.5 ± 1.1

<0.0001

HDL-cholesterol (mg/dl)

59.5 ± 2.1

56.7 ± 0.8

53.0 ± 0.4

46.9 ± 1.3

41.8 ± 2.5

<0.0001

LDL-cholesterol (mg/dl)

110.9 ± 25.2

110.3 ± 24.1

106.1 ± 24.3

106.5 ± 26.8

101.8 ± 28.2

>0.05

Triglyceride (mg/dl)

172.5 ± 145.7

170.6 ± 112.8

164.4 ± 100.0

193.4 ± 137.7

197.7 ± 137.5

>0.05

Uric acid (mg/dl)

4.5 ± 1.1

5.0 ± 1.0

5.5 ± 1.1

6.5 ± 1.1

7.4 ± 1.0

<0.0001

Creatinine (mg/dl)

0.48 ± 0.02

0.64 ± 0.06

0.89 ± 0.13

1.63 ± 0.2

3.6 ± 1.0

<0.0001

eGFR (ml/min/1.73 m2)

99.6 ± 10.5

70.4 ± 8.1

48.4 ± 7.6

24.0 ± 4.2

10.1 ± 2.7

<0.0001

Hemoglobin (g/dl)

13.5 ± 1.2 (n = 32)

14.0 ± 1.4 (n = 251)

14.0 ± 1.2 (n = 980)

12.3 ± 1.7 (n = 98)

10.1 ± 1.2 (n = 27)

<0.0001

Prevalence

 

 Smoking history (%)

11 (28.2)

143 (52.5)

729 (69.2)

71 (70.2)

20 (68.9)

<0.0001

 Hypertension (%)

32 (82.0)

207 (76.1)

619 (58.8)

86 (85.1)

28 (96.5)

>0.05

 Hyperlipidemia (%)

30 (76.9)

163 (59.9)

595 (56.5)

76 (75.2)

17 (58.6)

>0.05

 Hyperuricemia (%)

3 (7.6)

30 (11.0)

227 (21.5)

69 (68.3)

25 (86.2)

<0.0001

 Retinopathy (%)

22 (56.4)

142 (52.2)

478 (45.4)

78 (77.2)

26 (89.6)

<0.01

 CHD (%)

1 (2.56)

6 (2.21)

105 (9.98)

32 (31.68)

13 (44.83)

<0.0001

 Stroke (%)

2 (5.13)

11 (4.04)

87 (8.27)

28 (27.72)

6 (20.69)

<0.0001

 ASO (%)

1 (2.56)

4 (1.47)

32 (3.04)

16 (15.84)

15 (17.24)

<0.0001

 MVD (%)

4 (10.26)

19 (6.99)

187 (17.78)

53 (52.48)

16 (55.17)

<0.0001

Hypertension was defined as systolic blood pressure ≥130 mmHg and/or diastolic blood pressure ≥80 mmHg and/or current use of antihypertensive medication

Hyperlipidemia was defined as LDL-C level ≥120 mg/dl and/or current use of antihyperlipidemic medication

Hyperuricemia was defined as serum uric acid level ≥7.0 mg/dl in male, 6.0 mg/dl in female, and/or current use of antihyperuricemic medication

CHD was defined as a previous history of myocardial infarction or angina pectoris, confirmed by coronary interventions

Stroke was defined as a previous history of bleeding or ischemic stroke including lacuna infarction, confirmed by brain CT or MRI

ASO was diagnosed by angiography including enhanced CT or MRI and/or ankle-brachial pressure index (ABI) <0.9

MVD was defined as having one of the vascular diseases (CHD or stroke or ASO)

Statistical significance was estimated by a Cochran–Armitage test

CHD coronary heart disease, ASO arteriosclerosis obliterans, MVD macrovascular disease

The prevalence of concurrent MVD stratified by CKD stage is shown in Table 2 and Fig. 1. The prevalence of CHD was 2.56% in stage 1, 2.21% in stage 2, 9.98% in stage 3, 31.68% in stage 4, and 44.83% in stage 5. Stroke was 5.13% in stage 1, 4.04% in stage 2, 8.27% in stage 3, 27.72% in stage 4, and 20.69% in stage 5. The prevalence of ASO was 2.56% in stage 1, 1.47% in stage 2, 3.04% in stage 3, 15.84% in stage 4, and 17.24% in stage 5. MVD had a prevalence of 10.26% in stage 1, 6.99% in stage 2, 17.78% in stage 3, 52.48% in stage 4, and 55.17% in stage 5, showing a significant increase with the progression of CKD stage. The prevalence of CHD, stroke, ASO and MVD significantly increased with the progression of CKD stage (Cochran–Armitage test for trend: p < 0.0001). An analysis between consecutive stages from 2 to 3 and 3 to 4 showed significant increases (Ryan method: p = 0.00001, p = 0.000001) in MVD. Moving from stage 3 to 4 showed the most clinically significant increase.
https://static-content.springer.com/image/art%3A10.1007%2Fs10157-011-0420-6/MediaObjects/10157_2011_420_Fig1_HTML.gif
Fig. 1

Prevalence of vascular complications classified by CKD stages. Statistical significance was estimated by a Cochran–Armitage test for trend and by the Ryan method. CHD coronary heart disease, ASO arteriosclerosis obliterans, MVD macrovascular disease

To clarify the critical level of eGFR that predicts MVD prevalence, we used the ROC curve analysis. In terms of respective CKD stage, CKD stages 2 and 3 were significant in MVD prevalence (stage 2: p = 0.04, cut-off eGFR value of 66.2 ml/min/1.73 m2, area under the curve (AUC) 0.64; stage 3: p < 0.0001, cut-off eGFR value of 46.4 ml/min/1.73 m2, AUC 0.65). Other CKD stages (1, 4, and 5) were not significant. Although stage 2 was significant, the cut-off eGFR value of 66.2 ml/min/1.73 m2 was adjacent to the border of CKD stage 3. Figure 2 presents the ROC curve showing the association between the presence of MVD and eGFR in 1052 patients with stage 3 CKD.
https://static-content.springer.com/image/art%3A10.1007%2Fs10157-011-0420-6/MediaObjects/10157_2011_420_Fig2_HTML.gif
Fig. 2

Receiver operating characteristic curves for eGFR to predict MVD prevalence

The odds ratios for the risk of MVD in the eGFR range of ≥30 to <46 ml/min/1.73 m2 compared with the eGFR range of ≥46 to <60 ml/min/1.73 m2 was 2.47.

As a result, CKD stage 3 was classified into two substages by the prevalence of MVD. The risk factor for MVD in all patients was analyzed by multivariate logistic analysis (Table 3). Diabetes duration (p < 0.0001, odds ratio 1.05), proteinuria (p < 0.0001, odds ratio 1.93), eGFR <60 ml/min/1.73 m2 (p < 0.0001, odds ratio 2.92) and hyperuricemia (p = 0.0012, odds ratio 1.69) were significant independent risk factors after adjusting for various confounding factors. Proteinuria, eGFR <60 ml/min/1.73 m2 and hyperuricemia showed the highest odds ratios and were considered to be independent factors for MVD risk.
Table 3

Multiple logistic regression analysis for risk of MVD

Variable

Odds ratio

n = 1493 (male:female, 1273:220)

95%CI

p value

Gender (M)

0.690

0.448–1.063

0.09

Diabetes duration (years)

1.054

1.039–1.070

<0.0001

Smoking history

1.120

0.810–1.547

0.493

Body mass index (kg/m2)

0.990

0.953–1.029

0.623

Hypertension

1.221

0.888–1.679

0.2189

Hyperlipidemia

1.293

0.966–1.731

0.0846

Hyperuricemia

1.699

1.232–2.343

0.0012

Hemoglobin A1C (%)

0.929

0.815–1.060

0.2743

Proteinuria

1.933

1.406–2.658

<0.0001

eGFR < 60 ml/min/1.73 m2

2.925

1.752–4.883

<0.0001

Discussion

In Japan, the prevalence of diabetic nephropathy is increasing annually. Diabetic nephropathy ranks first in the annual number of new dialysis cases initiated. The survival outcome is unfavorable in comparison with other renal diseases. Early diagnosis and treatment of nephropathy and cardiovascular disease is essential to avoid the initiation of dialysis and to improve survival rates. Although CKD has been defined and classified, the management of individual kidney disease cases, especially diabetic nephropathy, is important.

CKD is an independent risk factor for cardiovascular disease, and mortality due to cardiovascular disease increases with the progression of CKD [24]. The results of our study suggest that as diabetic CKD progresses, the prevalence of CHD, stroke, ASO and MVD also increases. Therefore, diagnosis and therapeutic management are important, especially up to CKD stage 3. Furthermore, the possibility of a cardiovascular event occurring during the 3 years after a myocardial infarction also increases with advances in the CKD stage [4].

Diabetic patients possess risk factors for cardiovascular disease even before the onset of CKD, and the incidence increases after the onset of renal disease. The incidence and mortality of cardiovascular disease have been reported to increase after CKD stage 3 [7, 8]. Yamamoto et al. [9] studied 309 patients with CKD associated with type 2 diabetes (mean age 70.3 ± 9.5 years, diabetes duration 13.9 ± 7 years, 193 males and 116 females, eGFR 62.7 ± 9.8 ml/min/1.73 m2) and found that the prevalence of cardiovascular disease (7.5% in CKD stage 1, 11.8% in stage 2, 25% in stage 3, and 25% in stage 4 and higher) and stroke (7.5% in CKD stage 1, 17.6% in stage 2, 17% in stage 3, and 25% in stage 4 and higher) increased with the progression of the CKD stage. Their results were consistent with our present findings that the prevalence of CHD (2.56% in stage 1, 2.21% in stage 2, 9.98% in stage 3, 31.68% in stage 4, and 44.83% in stage 5) and stroke (5.13% in stage 1, 4.04% in stage 2, 8.27% in stage 3, 27.72% in stage 4, and 20.69% in stage 5) increased with the advance in CKD stage. In addition, the prevalence of these vascular diseases significantly increased as CKD progressed from stage 3 to 4. In our study, we found that the prevalence of ASO also followed the same trend. A recent report has indicated a correlation between diabetic nephropathy and the development of ASO; in particular, an eGFR less than 60 is an independent risk factor in men [10].

In diabetic and nondiabetic patients, the prevalence of cardiovascular disease was different. The incidence of cardiovascular disease in diabetic men has been reported to be twice as high as in nondiabetic men, and the incidence in diabetic women is three times higher than in nondiabetic women [11]. Another report indicates that the incidence of multivessel disease in acute myocardial infarction patients is higher in diabetic patients (66%) than in nondiabetic patients (46%) [12]. Furthermore, the United Kingdom Prospective Diabetes Study (UKPDS) reported that the cardiovascular mortality was 0.7% in subjects with no nephropathy, 2.0% in those with microalbuminuria, 3.5% in those with macroalbuminuria, and 12.1% in patients with elevated plasma creatinine level [13]. The Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines also state that the risk of cardiovascular disease increases in the presence of renal disease in diabetic patients [1]. In diabetic nephropathy, the risk of cardiovascular disease is higher than in nondiabetic nephropathy. Regarding the relationship between type 1 and type 2 diabetic nephropathy and cardiovascular disease, one study reported no difference in cardiovascular mortality [14]. On the other hand, another report indicated that compared with type 2 diabetic patients, type 1 diabetic patients are more susceptible to developing microvascular diseases but less likely to have concurrent coronary artery disease (myocardial infarction and heart failure) [15].

In the management of type 2 diabetic nephropathy, early diagnosis and clarifying the risk factors for cardiovascular disease is most important.

In our study, the prevalence of all MVDs increased from CKD stage 3 onward, while significant increases were found from stage 3 to 4. Based on our analysis of MVD prevalence according to the eGFR level, diabetic patients with an eGFR less than 46 ml/min/1.73 m2 had a significantly higher risk than those with an eGFR over 46 ml/min/1.73 m2. For early finding of MVD, including cardiovascular disease, the eGFR range of ≥30 to <46 ml/min/1.73 m2 is an important level in the diabetic CKD stage. CKD stage 3 should be divided into two substages for management of type 2 diabetic CKD from the perspective of cardiorenal association.

An association between cardiovascular risk and eGFR level was found in some reports.

Yamamoto et al. [9] observed a significantly higher prevalence of cardiovascular disease and stroke when the eGFR was lower than 75 ml/min in diabetic patients, including those without CKD. In the study by Nakagawa et al. [16], a comparison of the eGFR with pulse wave velocity (PWV) showed that an eGFR lower than 45 ml/min/1.73 m2 predicted arterial stiffness and progression to end-stage renal failure in nondiabetic CKD patients. Go et al. [3] reported that the risks of various outcomes, including hospitalization, death and cardiovascular disease, were increased with an eGFR below 60 ml/min/1.73 m2; however, these risks were not the same in stage 3 CKD patients. Furthermore, they divided CKD stage 3 patients (with or without diabetes) into two substages and examined the rates of death, cardiovascular events, and hospitalization. Their data showed an increased risk with an eGFR below 45 ml/min/1.73 m2 (hazard ratio for cardiovascular disease: 2.0 with an eGFR of 30–44 ml/min/1.73 m2 vs. 1.4 with an eGFR of 45–59 ml/min/1.73 m2). Go et al. did not consider underlying kidney disease. Our data imply that an eGFR of 46 ml/min/1.73 m2 is the cut-off for the presence or development of MVD, including CHD, stroke and ASO, in diabetic CKD patients. This eGFR level in diabetic CKD patients was very similar to Go et al.

Hypertension, hyperlipidemia, diabetes duration and CKD are known risk factors for cardiovascular disease. Our present study identified diabetes duration, proteinuria and hyperuricemia as independent risk factors for MVD. Of these factors, hyperuricemia (odds ratio: 1.69) and proteinuria (odds ratio: 1.93) showed the highest odds ratios and were considered to be the independent risk factors for MVD. Proteinuria is the conventional risk factor for cardiovascular disease in diabetic patients [13].

Hypertension was not a risk factor in our study because good control of blood pressure (130/73 mmHg) was being practiced by these patients. Our study showed that hyperuricemia was a risk factor. Recently, the association between uric acid and cardiovascular disease in type 2 diabetic patients was reported. Zoppini et al. [17] reported that elevated uric acid in type 2 diabetic patients is an independent risk factor for cardiovascular disease. Additionally, uric acid has been reported to be associated with the onset of type 2 diabetes [18] and progression of renal impairment in type 2 diabetic nephropathy [19, 20]. Madero et al. [21] reported that in patients with stages 3 to 4 CKD, hyperuricemia is associated with cardiovascular mortality, but not progression to kidney failure. The PIUMA study that followed patients with untreated essential hypertension for 4 years found that when the uric acid level is elevated, the risks of cardiovascular event, cardiovascular death and all-cause death were increased [22].

Uric acid level is conventionally considered to be a marker of renal disease because of the lowered uric acid excretion as CKD progresses; however, in a rat model of hyperuricemia, hypertension and renal arteriolopathy are also observed, which can be suppressed by antihyperuricemic agents [2325]. These reports suggest that uric acid is a true mediator of progression and aggravation of renal impairment. The importance of uric acid management in diabetic patients so as to retard the progression of diabetic CKD as well as prevent cardiovascular disease development should be clarified.

Our study concluded that the prevalence of MVD, including CHD, stroke and ASO, increases with the progression of CKD stage in type 2 diabetic CKD patients, and this trend is especially seen as CKD progresses from stage 3 to 4. In stage 3 CKD patients, an eGFR of 46 ml/min/1.73 m2 predicted MVD prevalence. CKD stage 3 should be divided into two substages (eGFR of ≥30 to <46 and ≥46 to < 60 ml/min/1.73 m2). From the present study, an eGFR of 46 ml/min/1.73 m2 is a useful critical level marker for the prevention and management of type 2 diabetic CKD from the perspective of cardiorenal association. Hyperuricemia is an independent risk factor for MVD, suggesting that uric acid management is important to prevent cardiorenal association in type 2 diabetic CKD.

Conflict of interest

The authors have no conflict of interest to disclose.

Copyright information

© Japanese Society of Nephrology 2011