Relationship between peripheral artery disease and combined albuminuria and low estimated glomerular filtration rate among elderly patients with type 2 diabetes mellitus

Abstract

This study aimed at investigating the combined effects of albuminuria and reduced estimated glomerular filtration rate (eGFR) on peripheral artery disease (PAD) among elderly patients with diabetes. A total of 236 subjects were cross-classified into four groups according to the presence or absence of albuminuria (urinary albumin creatinine ratio (ACR) ≥ 30 mg/g) and low eGFR (<60 mL/min/1.73 m²). Cardiovascular risk factors and the ankle–brachial index (ABI) were also assessed. After multivariate adjustment using logistic regression analysis, the odds ratios (OR) for prevalent PAD related to albuminuria with preserved eGFR, normoalbuminuria with low eGFR and albuminuria with low eGFR compared to normoalbuminuria with preserved eGFR were 1.10 [95% confidence interval (CI) = 0.43–2.79], 3.14 (95% CI = 1.20–8.22) and 3.87 (95% CI = 1.72–8.72), respectively. In conclusion, in elderly patients with type 2 diabetes, both normoalbuminuria with low eGFR and albuminuria with low eGFR are associated independently with PAD.

Keywords

Albuminuria low estimated glomerular filtration rate elderly patients type 2 diabetes mellitus

Introduction

Peripheral artery disease (PAD) is a diffuse atherosclerotic vascular disease commonly present in patients with diabetes, particularly for the elderly patients aged more than 65 years.¹ In Taiwan, its respective prevalence for the age groups of 65–74 and ≥75 years is 12.9% and 25.3%, respectively.² PAD has been linked with cardiovascular disease (CVD) as well as the risk of occurrence of foot-related injuries and amputations.^3,4 It has been confirmed that elderly patients with type 2 diabetes after lower extremity amputations are more likely to suffer from a 1.9-fold higher risk of mortality if compared with those of younger patients.⁵

Epidemiological studies have identified several risk factors that are associated independently with PAD, among which are older age, smoking habit, higher triglyceride concentration, higher systolic blood pressure (SBP) and pulse pressure, elevated high-sensitive c-reactive peptide (hs-CRP), chronic kidney disease (CKD), presence of albuminuria and low estimated glomerular filtration rate (eGFR).^6–8 However, among these factors, the role of both albuminuria and low eGFR in the development of PAD among patients with diabetes has not been well elucidated. Previous studies have indicated that albuminuria is significantly correlated with PAD in elderly patients with diabetes.^8,9 In contrast, a recent investigation established that reduced eGFR, independent of albuminuria, is strongly related to PAD in subjects with diabetes. It was proposed that as diabetes progressed, the predictive value of albuminuria gradually diminished after adjustment for renal function.¹⁰

In fact, these investigations only relied on a single criterion to define the risk of PAD, either the presence of albuminuria or a reduction in the eGFR, and their combined effects of both metrics were not evaluated simultaneously.^9–11 Since albuminuria and reduced eGFR might be closely related, the correlation between albuminuria and PAD could be confounded by renal function. Consequently, the results of previous research might have underestimated the magnitude of their association.¹² On the basis of these findings, both criteria of CKD should be used simultaneously for the purpose to better characterize the relationship between kidney disease and PAD.

Therefore, this study aimed first to investigate the characteristics of subjects when stratified with PAD and without PAD and second to determine the association between the combined effect of albuminuria and low eGFR with the presence of subclinical PAD in elderly patients with type 2 diabetes mellitus.

Methods

Study population

A total of 236 patients with type 2 diabetes mellitus, aged ≥65 years, were enrolled consecutively from the outpatient department of endocrinology and metabolism at a regional hospital in southern Taiwan. Study patients were mostly recruited into the diabetes shared care programme (DSCP), which provided appropriate care to the population with diabetes. The programme aimed at reducing the incidence of diabetes complications including PAD via a comprehensive strategy of screening and early intervention of the risk factors targeting hyperglycaemic, dyslipidemia and hypertension. Among the recruited patients, diagnosis of type 2 diabetes mellitus was according to the criteria of the American Diabetes Association (ADA), and patients were treated with either oral hypoglycaemic agents or insulin at the time of recruitment.¹³ Subjects with acute kidney disease, moderate-to-severe congestive heart failure (New York Heart Association classification of III or IV), fever status and urinary tract infection were excluded. The protocol for this observation study was approved by our Institutional Review Board, and all participants provided informed consent.

Clinical and laboratory measurements

Demographic and medical data, including age, gender, history of smoking, duration of diabetes and comorbid conditions such as CVD and retinopathy, were garnered from medical records. The presence of CVD was diagnosed by the physician as a history of stroke or coronary heart disease (CHD). CHD was defined if they had a history of typical angina, the presence of electrocardiogram (ECG) abnormalities suggestive of CHD, old myocardial infarction or they had undergone coronary artery bypass surgery or angioplasty. Stroke was defined if they had a history of cerebrovascular incidents such as cerebral bleeding or infarction. Retinopathy was assessed by fundoscopy after pupil dilatation. History of smoking was defined as never/past/current.

Central obesity was defined if abdominal circumference was ≥90 cm in men and ≥80 cm in women. Blood pressure was measured by trained observers using an automatic oscillometric monitor while the patients were in the sitting position. Otherwise, information regarding patient medications, including anti-hypertensive agent, statin/fibrate, anti-platelet agent and insulin during the study period, was obtained from medical records.

All laboratory measurements were performed by the Yuan’s General Hospital laboratory using standardized and automated methods. Glycated haemoglobin (HbA1c) was determined by an ion-exchange high-performance liquid chromatography (HLC-723 G8; Tosoh, Tokyo, Japan). Serum creatinine and lipid profiles, including measurements of total cholesterol, triglyceride, high-density lipoprotein (HDL) and low-density lipoprotein (LDL), were measured on an Olympus AU2700 autoanalyzer (Beckman Coulter, Mishima, Japan) using an enzymatic assay. Hs-CRP was measured using an immunonephelometry assay (BN II; Siemens, Deerfield, IL, USA). In addition, urinary creatinine concentration and urinary albumin were measured on an Olympus AU2700 autoanalyzer using the alkaline picrate-kinetic method and immunoturbidimetric assay, respectively.

Measurement of albuminuria and low GFR

The urinary albumin excretion rate was presented as the albumin creatinine ratio (ACR), which was measured from an early morning spot urine sample. Albuminuria was defined as an ACR ≥ 30 mg/g based on the recommendation of ADA.¹⁴ The eGFR was estimated using the Modification of Diet in Renal Disease (MDRD) Study formula as follows: 186 × [serum creatinine (mg/dL)]^−1.154 × (age)^−0.203 × (0.742 if female).¹⁵ Preserved eGFR was defined as an eGFR of ≥60 and low eGFR or renal insufficiency was defined as an eGFR of <60.

In this study, subjects were then cross-classified into four distinct subgroups determined by albuminuria and low eGFR, defined as follows: normoalbuminuria with preserved eGFR, albuminuria with preserved eGFR, normoalbuminuria with low eGFR and albuminuria with low eGFR.

Ankle–brachial index measurements

In an effort to reduce elderly participant burden, only one measurement was taken for all participants. In brief, a bidirectional portable echo-Doppler of 8 MHz (VasoGuard MicroLite; Nicolet, Madison, WI, USA) and a calibrated mercury sphygmomanometer were used to measure simultaneously bilateral brachial and ankle SBPs. For this measurement, patients were in the supine position in a quiet room after a 10-min rest. The device automatically calculated ankle–brachial index (ABI) by dividing the ankle systolic pressure of either side by the highest of the right and left brachial systolic pressures. Subjects who had an ABI < 0.9 or > 1.4 in either leg were defined as having PAD, which was based on the results of a recent review.¹⁶ If both ABI measures were ≥0.9 and ≤1.4, subjects were defined as having normal ABI.

Statistical analysis

Quantitative variables are presented as mean ± standard deviation or median (interquartile range) if they were normally or non-normally distributed. Qualitative variables are expressed as absolute numbers and percentages. The study groups stratified by prevalent PAD were compared on clinical characteristics using chi-square test for categorical variables, independent t-test for continuous variables and Mann–Whitney test for ACR and hs-CRP.

The odds ratios (OR) for prevalent PAD associated with the cross-classification of albuminuria and low eGFR were calculated using logistic regression models, with subjects having normoalbuminuria with preserved eGFR serving as the reference category. Initially, Model 1 had only minimal adjustment including age and gender. A subsequent fully multivariable-adjusted Model 2 included additional adjustment for those factors that were statistically significant at α < 0.05 in univariate logistic regression analysis (data not shown). Given our sample sizes, our study was powered to detect an estimated OR comparing PAD with non-PAD of 3.09 for albuminuria with eGFR ≥ 60, 3.62 for normoalbuminuria with eGFR < 60 and 2.94 for albuminuria with eGFR < 60, setting at 80% power. The statistical operations were performed using SPSS software (SPSS Version 19; SPSS Inc., Chicago, IL, USA), and the power analysis was carried out with NCSS 2007. A p value of less than 0.05 was considered significant.

Results

This study analysed 236 elderly subjects (47.9% men) with an average age of 72.0 ± 5.0 years. The mean ABI of the overall population was 0.98 ± 0.22; 26.7% of the participants had an ABI < 0.9 or > 1.4, 13.6% had an ABI between 0.7 and 0.89, 8.1% had an ABI between 0.4 and 0.69, 2.1% had an ABI < 0.4 and only 3% had an ABI > 1.4. Regarding the subjects with symptomatic PAD, 12 (5.1%) had intermittent claudication, 4 (1.7%) had foot lesion and 2 (0.9%) had lower extremity amputation.

These study subjects were categorized into PAD and without PAD, and their clinical characteristics were compared (Table 1). Patients with PAD were characterized by more CVD; longer duration of diabetes; higher total cholesterol, triglyceride, LDL-cholesterol, hs-CRP and ACR; and lower HDL-cholesterol and eGFR. Additionally, they had higher prevalence of normoalbuminuria with eGFR < 60 and albuminuria with eGFR < 60, which is twofold higher if compared with subjects without PAD.

Table 1.

Comparison of clinical characteristics according to prevalent peripheral artery disease.

Characteristics	Peripheral artery disease		p value
Characteristics	No, 0.9 ≤ ABI ≤ 1.4 (n = 169)	Yes, ABI < 0.9 or > 1.4 (n = 67)	p value
Age, years	71.7 ± 4.9	72.8 ± 5.2	0.116
Male gender, n (%)	85 (50.3)	28 (41.8)	0.238
Cardiovascular disease, n (%)	15 (8.9)	13 (19.4)	0.024*
Retinopathy, n (%)	104 (65.8)	32 (53.3)	0.089
Smoking, n (%)			0.550
Never	129 (76.3)	50 (74.6)
Past	27 (16.0)	9 (13.4)
Current	13 (7.7)	8 (11.9)
Central obesity	114 (67.5)	48 (71.6)	0.532
Body mass index (kg/m²)	25.9 ± 3.8	26.0 ± 3.9	0.943
Duration of diabetes (years)	13.6 ± 8.7	16.5 ± 8.7	0.020*
Systolic blood pressure (mmHg)	138.5 ± 17.5	141.5 ± 15.3	0.216
Diastolic blood pressure (mmHg)	74.8 ± 10.3	74.8 ± 9.7	0.991
Haemoglobin A1c, % (mmol/mol)	7.43 ± 0.90 (57.7 ± 7.0)	7.44 ± 0.89 (57.8 ± 6.9)	0.952
Total cholesterol (mmol/L)	4.22 ± 0.73	4.46 ± 0.79	0.031*
Triglyceride (mmol/L)	1.48 ± 0.82	1.76 ± 0.97	0.024*
HDL-cholesterol (mmol/L)	1.20 ± 0.30	1.07 ± 0.27	0.002*
LDL-cholesterol (mmol/L)	2.25 ± 0.55	2.44 ± 0.52	0.017*
hs-CRP (mg/L)	0.09 (0.05–0.19)	0.16 (0.05–0.40)	0.016*
eGFR (mL/min/1.73 m²)	76.6 ± 26.8	55.5 ± 23.2	<0.001*
ACR (mg/g)	23.0 (7.6–81.9)	35.6 (10.1–250.8)	0.020*
Medication (%)
Anti-hypertensive agent	116 (69.0)	49 (73.1)	0.536
Statin/fibrate	86 (50.9)	36 (53.7)	0.693
Anti-platelet agent	62 (36.7)	29 (43.3)	0.348
Insulin	29 (17.2)	16 (23.9)	0.236
Cross-classification of albuminuria and eGFR, n (%)			<0.001*
Normoalbuminuria and eGFR ≥ 60	79 (46.7)	15 (22.4)
Albuminuria and eGFR ≥ 60	40 (23.7)	10 (14.9)
Normoalbuminuria and eGFR < 60	18 (10.7)	14 (20.9)
Albuminuria and eGFR < 60	32 (18.9)	28 (41.8)

ABI: ankle–brachial index; HDL: high-density lipoprotein; LDL: low-density lipoprotein; hs-CRP: high-sensitive c-reactive peptide; eGFR: estimated glomerular filtration rate; ACR: albumin creatinine ratio.

Data are presented as mean ± standard deviation or numbers (percentages), except for hs-CRP and ACR, which are presented as median (interquartile range).

p < 0.05.

As shown in Figure 1, the prevalence of PAD was greatest in those with albuminuria and low eGFR. Specifically, the prevalence of PAD was 16% among patients with normoalbuminuria and eGFR ≥ 60 and increased up to 20%, 43.8% and 46.7% in patients presenting as albuminuria with eGFR ≥ 60, normoalbuminuria with eGFR < 60 and albuminuria with eGFR < 60, respectively.

Figure 1.

Prevalence of peripheral artery disease after cross-classification of albuminuria and low eGFR.

In Table 2, we performed the multivariate analysis to assess the association between different combined albuminuria and low eGFR categories and PAD. After age and gender adjustment, and compared to participants with normoalbuminuria and eGFR ≥ 60, the ORs [95% confidence interval (CI)] of PAD were 1.28 (0.52–3.14), 3.92 (1.58–9.78) and 4.52 (2.11–9.68) for those with albuminuria and eGFR ≥ 60, normoalbuminuria and eGFR < 60 and albuminuria and eGFR < 60, respectively (Model 1). Finally, with additional adjustment for those factors that were statistically significant at α < 0.05 in univariate logistic analysis (CVD, duration of diabetes, total cholesterol, triglyceride), the OR and 95% CI became 1.10 (0.43–2.79), 3.14 (1.20–8.22) and 3.87 (1.72–8.72), respectively (Model 2). Taken together, adjustment for multiple risk factors for PAD did not result in much attenuation in the odds of PAD among patients presenting as normoalbuminuria with eGFR < 60 and albuminuria with eGFR < 60.

Table 2.

Minimal and multivariable-adjusted ORs (95% confidence interval) for the prevalent peripheral artery disease associated with cross-classification of albuminuria and low eGFR.

Cross-classification of albuminuria and eGFR	N	Crude OR	Model 1	Model 2
NA and eGFR ≥ 60	94	1 (Ref)	1 (Ref)	1 (Ref)
A and eGFR ≥ 60	50	1.32 (0.54–3.19)	1.28 (0.52–3.14)	1.10 (0.43–2.79)
NA and eGFR < 60	32	4.10 (1.68–9.98)*	3.92 (1.58–9.78)*	3.14 (1.20–8.22)*
A and eGFR < 60	60	4.61 (2.18–9.75)*	4.52 (2.11–9.68)*	3.87 (1.72–8.72)*

OR: odds ratio; eGFR: estimated glomerular filtration rate; NA: normoalbuminuria; A: albuminuria.

Model 1: adjusted for age and gender. Model 2: adjusted for age, gender and those variables that were statistically significant at α < 0.05 in univariate logistic regression analysis (cardiovascular disease, duration of diabetes, total cholesterol, triglyceride).

p < 0.05.

Discussion

The results of this cross-sectional study involving elderly patients with type 2 diabetes showed that the prevalence of PAD was 26.7% in the overall study population. After cross-classification by the presence of albuminuria and low eGFR, our data demonstrated that normoalbuminuric renal insufficiency and albuminuric renal insufficiency were associated significantly with threefold to fourfold increased likelihood of PAD. We suggest that these two criteria are distinct and complementary, and this might reflect pathophysiological differences among these different cross-classification renal stages.

In this study, we defined PAD as an ABI < 0.9 or > 1.4 based on the findings from several recent investigations instead of only ABI < 0.9.^10,16,17 In fact, high ABI values (>1.4) are observed in patients with diabetes, particularly for those concomitant with CKD.¹⁸ According to a subanalysis of the Strong Heart Study, using the 0.90 as the cut-off point may lead to underestimation of the prevalence of PAD because an increased ABI also reflects coronary calcifications and CVD risk.¹⁷ Furthermore, it also led to clinical manifestation of PAD and increased risk of amputation, proposing that elevation of this index could reflect diffuse atheromatous disease.^19,20 Since the manifestations or prognosis of high ABI (>1.4) is equivalent with those of low ABI (<0.9), subjects with elevated ABI are thus included in the PAD group for current analysis.

It is well known that most previous analyses have limited their classification to either albuminuria or reduced eGFR to investigate its association with prevalence of PAD.^7–10,21 Indeed, the presence of either albuminuria or a reduced eGFR identified different categories of the CKD because only 54.4% of those with albuminuria had a reduced eGFR and only 64.5% of those with a reduced eGFR had albuminuria as shown in our report. This suggested that relying on single criterion alone will miss many cases of pre-existing kidney disease. Prior studies had established that the presence of albuminuria was significantly associated with PAD.^9,11 On the contrary, a recent investigation found that reduced eGFR but not albuminuria was independently associated with PAD in patients with diabetes.¹⁰ For the research, which examined the combined effects of albuminuria and reduced eGFR on PAD, the result stated that the prevalence of PAD was significantly elevated accompanied by both abnormalities.⁶ A recent population-based study also indicated that the prevalence of PAD was greatest in those with both albuminuria and reduced eGFR, and this association was statistically significant in those with albuminuria alone or reduced eGFR alone.¹²

Compared with these previous studies, our research focused on elderly patients with type 2 diabetes. Interestingly, we found that albuminuria with preserved eGFR, which represented the early stage of diabetic nephropathy, was not significantly related to PAD. In contrast, among the patients with late-stage diabetic nephropathy, a significant association between PAD and low eGFR was observed. This finding was in line with the results of a previous report, which revealed that the prevalence of asymptomatic PAD increased with more advanced stage of CKD.²² To our knowledge, the presence of albuminuria has been recognized as a risk factor of generalized vascular damage. On the other hand, albuminuria may be just a marker of poor glucose control or longer duration of diabetes.^23,24 A recent large population-based study demonstrated that albuminuria was not associated with PAD in diabetic subjects. However, in further subgroup analysis, a possible albuminuria and PAD correlation was found in patients with intense glucose control.¹⁰ In summary, the severity of vascular damage cannot be decided simply by the pre-existence of albuminuria, particularly among patients with early-stage diabetic nephropathy.

Although the pathophysiologic mechanisms between PAD and both albuminuria and low eGFR are not fully elucidated, it is believed that the presence of both abnormalities are markers of the generalized burden of atherosclerotic disease.²⁵ In this study, deterioration of renal function instead of presence of albuminuria may reflect more accurately the severity of generalized atherosclerosis.²⁶ A prior study confirmed that atherosclerotic change in the renal vessels commonly occurs in the presence of PAD: 56% had either renal artery stenosis or occlusion.²⁷ Other factors such as hyperhomocysteinemia increase inflammation activity and oxidative stress and could also contribute to the higher prevalence of PAD in patients with renal insufficiency.^28,29 In contrast, albuminuria seems related to the generalized endothelial dysfunction, may also be a marker of a prothrombotic state and could possibly lead to the development of lower extremity atherosclerosis.³⁰ However, this insight mechanism warranted further investigations for confirmation.

The strength of this study is that we studied the combined effects of albuminuria and reduced eGFR, thus making it more likely to find valid associations between CKD and PAD. Several limitations of our study should also be noted. First, the cross-sectional nature limits conclusions about causality and directionality of the association observed in this study. Second, albuminuria measurement was based on a single spot urine sample. However, a prior study indicated that an early morning urinary ACR ≥ 30 mg/g is associated with an albumin excretion rate ≥ 30 mg/min, with high sensitivity and specificity.³¹ Third, misclassification due to transient albuminuria could not be completely excluded. Nevertheless, this misclassification would bias the observed results towards the null. Fourth, misclassification of PAD was possible because PAD was diagnosed by a non-invasive method rather than a gold standard technique such as angiography. However, angiography is rarely applied to the elderly patient for the purpose of diagnosis only.

In conclusion, we indicate that both normoalbuminuric and albuminuric renal insufficiencies are independently associated with the presence of PAD among elderly patients with type 2 diabetes mellitus, after adjustment of multiple CVD risk factors. This cross-classification method characterizes a subgroup of patients who have very high risk of PAD as well as CVD, and it is hoped that they would benefit from the early implementation of appropriate therapeutic strategies.

Footnotes

Acknowledgements

The authors thank Mrs Chih-Yun Lin for her excellent statistical work.

Declaration of conflicting interest

The authors report no conflicts of interest.

Funding

This study was supported by a grant from Yuan’s General Hospital (RG10-005), Taiwan.

References

Norman

Davis

Bruce

. Peripheral arterial disease and risk of cardiac death in type 2 diabetes: the Fremantle Diabetes Study. Diabetes Care 2006; 29: 575–580.

Tseng

. Prevalence and risk factors of peripheral arterial obstructive disease in Taiwanese type 2 diabetic patients. Angiology 2003; 54: 331–338.

Murabito

Evans

Nieto

. Prevalence and clinical correlates of peripheral arterial disease in the Framingham Offspring Study. Am Heart J 2002; 143: 961–965.

Davis

Norman

Bruce

. Predictors, consequences and costs of diabetes-related lower extremity amputation complicating type 2 diabetes. Diabetologia 2006; 49: 2634–2641.

Tseng

Chong

Tseng

. Mortality, causes of death and associated risk factors in a cohort of diabetic patients after lower-extremity amputation: a 6.5-year follow-up study in Taiwan. Atherosclerosis 2008; 197: 111–117.

Mostaza

Suarez

Manzano

. Relationship between ankle-brachial index and chronic kidney disease in hypertensive patients with no known cardiovascular disease. J Am Soc Nephrol 2006; 17: 201–205.

Subramaniam

Nang

Lim

. Distribution of ankle–brachial index and the risk factors of peripheral artery disease in a multi-ethnic Asian population. Vasc Med 2011; 16: 87–95.

Chou

Weng

Chang

. Analysis of traditional and nontraditional risk factors for peripheral arterial disease in elderly type 2 diabetic patients in Taiwan. Diabetes Res Clin Pract 2008; 81: 331–337.

Tseng

Chong

Tseng

. The association between urinary albumin excretion and ankle-brachial index in elderly Taiwanese patients with type 2 diabetes mellitus. Age Ageing 2008; 37: 77–82.

10.

Yang

Tsai

. Association of low glomerular filtration rate and albuminuria with peripheral arterial disease: the National Health and Nutrition Examination Survey, 1999–2004. Atherosclerosis 2010; 209: 230–234.

11.

Wattanakit

Folsom

Criqui

. Albuminuria and peripheral arterial disease: results from the Multi-Ethnic Study of Atherosclerosis (MESA). Atherosclerosis 2008; 201: 212–216.

12.

Baber

Mann

Shimbo

. Combined role of reduced estimated glomerular filtration rate and microalbuminuria on the prevalence of peripheral arterial disease. Am J Cardiol 2009; 104: 1446–1451.

13.

Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 2003; 26: 5–20.

14.

American Diabetes Association. Standards of medical care in diabetes – 2007. Diabetes Care 2007; 30: S4–S41.

15.

Levey

Bosch

Lewis

. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999; 130: 461–470.

16.

Potier

Abi Khalil

Mohammedi

. Use and utility of ankle brachial index in patients with diabetes. Eur J Vasc Endovasc Surg 2011; 41: 110–116.

17.

Resnick

Lindsay

McDermott

. Relationship of high and low ankle brachial index to all-cause and cardiovascular disease mortality: the strong heart study. Circulation 2004; 109: 733–739.

18.

Young

Adams

Anderson

. Medial arterial calcification in the feet of diabetic patients and matched non-diabetic control subjects. Diabetologia 1993; 36: 615–621.

19.

Everhart

Pettitt

Knowler

. Medial arterial calcification and its association with mortality and complications of diabetes. Diabetologia 1988; 31: 16–23.

20.

Tsuchiya

Suzuki

Egawa

. Abnormal peripheral circulation in type 2 diabetic patients with normal ankle-brachial index associates with coronary atherosclerosis, large artery stiffness, and peripheral vascular resistance. Diabetes Res Clin Pract 2005; 70: 253–262.

21.

Choi

Kim

Lee

. eGFR is associated with subclinical atherosclerosis independent of albuminuria: the Dong-gu Study. Atherosclerosis 2010; 212: 661–667.

22.

Ungprasert

Pornratanarangsi

. Correlation between peripheral arterial disease and stage of chronic kidney disease. J Med Assoc Thai 2011; 94: S46–S50.

23.

Nieuwdorp

Mooij

Kroon

. Endothelial glycocalyx damage coincides with microalbuminuria in type 1 diabetes. Diabetes 2006; 55: 1127–1132.

24.

Krolewski

Laffel

Krolewski

. Glycosylated hemoglobin and the risk of microalbuminuria in patients with insulin-dependent diabetes mellitus. N Engl J Med 1995; 332: 1251–1255.

25.

Kramer

Nguyen

Curhan

. Renal insufficiency in the absence of albuminuria and retinopathy among adults with type 2 diabetes mellitus. JAMA 2003; 289: 3273–3277.

26.

Mohler

III . Peripheral arterial disease: identification and implications. Arch Intern Med 2003; 163: 2306–2314.

27.

lra

Guo

Kausz

. Atherosclerotic renovascular disease in United States patients aged 67 years or older: risk factors, revascularization, and prognosis. Kidney Int 2005; 68: 293–301.

28.

Nihei

Tasaki

Yamashita

. Hyperhomocysteinemia is associated with human coronary atherosclerosis through the reduction of the ratio of endothelium-bound to basal extracellular superoxide dismutase. Circ J 2004; 68: 822–828.

29.

Knight

Rimm

Pai

. Kidney dysfunction, inflammation, and coronary events: a prospective study. J Am Soc Nephrol 2004; 15: 1897–1903.

30.

Kario

Matsuo

Kobayashi

. Activation of tissue factor-induced coagulation and endothelial cell dysfunction in non-insulin-dependent diabetic patients with microalbuminuria. Arterioscler Thromb Vasc Biol 1995; 15: 1114–1120.

31.

Hutchison

O’Reilly

MacCuish

. Albumin excretion rate, albumin concentration, and albumin/creatinine ratio compared for screening diabetics for slight albuminuria. Clin Chem 1988; 34: 2019–2021.