- Email: [email protected]
Endothelial vasomotor dysfunction in the brachial artery predicts the short-term development of early stage renal dysfunction in patients with coronary artery disease
International Journal of Cardiology 148 (2011) 183–188
Contents lists available at ScienceDirect
International Journal of Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j c a r d
Endothelial vasomotor dysfunction in the brachial artery predicts the short-term development of early stage renal dysfunction in patients with coronary artery disease Takamitsu Nakamura, Jun-ei Obata, Mitsumasa Hirano, Yoshinobu Kitta, Keita Sano, Tsuyoshi Kobayashi, Daisuke Fujioka, Yukio Saito, Toshiaki Yano, Kenichi Kawabata, Kazuhiro Watanabe, Yosuke Watanabe, Hideto Mishina, Kiyotaka Kugiyama ⁎ University of Yamanashi, Faculty of Medicine, Department of Internal Medicine II, Chuo, Yamanashi, Japan
a r t i c l e
i n f o
Article history: Received 23 July 2009 Accepted 31 October 2009 Available online 27 November 2009 Keywords: Endothelial vasomotor function Renal failure Predictor Coronary artery disease
a b s t r a c t Background: This study examined whether endothelial vasomotor dysfunction in the brachial artery predicted early renal dysfunction in patients with coronary artery disease (CAD). Endothelial function in the renal vasculature plays an important role in the regulation of renal hemodynamics. As endothelial dysfunction is a systemic disorder, there may be a relationship between endothelial function in the brachial artery and renal vasculature. Methods: Flow-mediated endothelium-dependent dilation (FMD) in brachial artery and renal functional parameters were measured in 757 patients with CAD without macroalbuminuria. Results: In a cross-sectional data, an impaired FMD was associated with higher serum creatinine levels and urinary albumin excretion (UAE), lower creatinine clearance rate and estimated glomerular filtration rate (eGFR) at baseline in multiple linear regression analysis. In a follow-up study including a subgroup of 448 patients with normal renal function (serum creatinine level b1.0 mg/dL, UAE b 25 mg/day and eGFR ≥ 60 mL/ min/1.73 m2 at baseline), 96 patients had an endpoint of early stage renal dysfunction (serum creatinine levels ≥ 1.2 mg/dL, UAE ≥ 30 mg/day and/or eGFR b 60 mL/min/1.73 m2) during 12 month follow-up. Multivariate logistic regression analysis showed that impaired FMD was significantly associated with progression to the early stage renal dysfunction after adjustment with age, diabetes mellitus, hypertension and C-reactive protein levels. Conclusions: Endothelial vasomotor dysfunction in the brachial artery is independently associated with progression from normal renal function to early stage renal dysfunction in patients with CAD. Measurement of FMD may therefore be useful for assessing risk of future renal dysfunction. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Chronic kidney disease is associated with a high prevalence of cardiovascular events [1,2]. There is evidence that treatment at the earliest possible stage of renal dysfunction is more effective for preventing progression to end-stage renal disease and subsequent cardiovascular disease [3–7]. It is therefore clinically important to detect this group of patients who are at high risk of developing early stage renal disease [7]. Microalbuminuria, a marker of early stage renal dysfunction, is related to renal microvasculopathy, characterized by preglomerular arteriolar involvement and tubulointerstitial
⁎ Corresponding author. Department of Internal Medicine II, University of Yamanashi, Faculty of Medicine, 1110 Shimokato, Chuo 409-3898, Japan. Tel.: + 81 55 273 9590; fax: + 81 55 273 6749. E-mail address: [email protected] (K. Kugiyama). 0167-5273/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2009.10.054
changes [1,2,8]. Experimental studies indicate that endothelial dysfunction in the renal vasculature is involved in the development of renal microvasculopathies and microalbuminuria [9–11]. As endothelial dysfunction is a systemic disorder [12], it is possible that there is a relationship between endothelial dysfunction in the systemic arteries and renal vasculature [13–15]. A previous crosssectional study showed that an impaired endothelial vasomotor response in the forearm was associated with mild to moderate renal dysfunction in hypertensive patients without previous cardiovascular diseases [16]. In contrast, in vitro measurement of endotheliumdependent dilation in isolated internal thoracic arteries showed no significant relationship with the presence of mild to moderate renal dysfunction in patients with coronary artery disease (CAD) [17]. Therefore, it remains controversial whether there is an association between endothelial vasomotor dysfunction in systemic arteries and renal dysfunction in patients with cardiovascular disease and advanced atherosclerotic burden. It also remains unknown whether
184
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
endothelial vasomotor dysfunction in systemic arteries precedes renal dysfunction, thereby providing a predictive value of future development of renal dysfunction. In the present study, we examined whether endothelial vasomotor dysfunction in the brachial artery provided prognostic information on the development of early stage renal dysfunction in patients with CAD that is linked with chronic kidney disease. 2. Methods 2.1. Study patients This study included 757 consecutively enrolled patients with stable CAD without macroalbuminuria (≥300 mg/day) at Yamanashi University Hospital from January 1, 2002 to Dec 31, 2006. All the patients had a routine screen following enrollment in the study that included measurement of flow-dependent dilation (FMD) in the brachial artery using Bmode ultrasound images, serum creatinine levels, 24 h urine albumin excretion (UAE) and calculation of creatinine clearance (CCr) and estimated glomerular filtration rate (eGFR). All patients had angiographic evidence of organic diameter stenosis N 70% of at least one major coronary artery (single-vessel disease n = 175; two-vessel disease n = 302; and three-vessel disease n = 280). The exclusion criteria were: 1) a history of dialysis, 2) use of contrast media 3 months prior to enrollment, 3) congestive heart failure (≥New York Heart Association classification III), 4) secondary hypertension, 5) renal artery stenosis, 6) polycystic kidney disease, 7) chronic inflammatory diseases, 8) major injury or surgery 3 months prior to enrollment, and 9) other serious diseases. The baseline characteristics of the patients in the study are summarized in Table 1. All the patients gave written, informed consent for the study at enrollment and received the standardized medications outlined in Table 1. This study was conducted in accordance with the guidelines approved by the ethics committee of Yamanashi University Hospital and conformed with the principles outlined in the 1975 Declaration of Helsinki. 2.2. Measurements of renal function Creatinine levels in serum and urine were measured by an enzymatic method using an auto-analyzer (Mitsubishi Kagaku Iatron, Tokyo, Japan). Serum creatinine levels were measured on 2 days a week apart, with the average value being used in the analyses. CCr was calculated from 24 h urine creatinine concentration and the serum creatinine concentration on the same day, and the value normalized to a surface area of 1.73 m2 [6]. GFR was estimated by Modification of Diet in Renal Disease study equation [1]. Urinary albumin levels were measured by immunonephelometry, and fasting plasma high sensitivity C-reactive protein (CRP) levels by rate nephelometry (Dade Behring, Marburg, Germany). 2.3. Measurements of endothelial vasomotor function of the brachial artery assessed by flow-mediated dilation (FMD) Vasodilator responses in the brachial arteries were measured using B-mode ultrasound images with a 7.5-MHz linear array transducer (HP-5500, Phillips Corp., Tokyo, Japan) according to a method validated in our previous studies [18,19]. Briefly, after baseline measurements of the diameter and flow velocity in the brachial artery, a blood pressure cuff was placed around the forearm and inflated to a pressure of 250 to 300 mm Hg for 5 min and then released. Measurements of the artery diameter during the reactive hyperemia were taken 45 to 90 s after cuff deflation. Sublingual nitroglycerin (300 μg) was then administered and the measurements repeated 3 min later. The responses of the vessel diameters to reactive hyperemia and nitroglycerin were expressed as the percentage increase in diameter from the baseline value. The diameters of the vessel responses were assessed at three points along a 10-mm length of the artery, and the values averaged. Blood flow was calculated by multiplying the velocity–time integral of the Doppler flow signal by heart rate and cross-sectional area of the vessel. The increase in brachial blood flow was calculated as the maximum flow recorded in the first 15 s after cuff deflation and was expressed as a percentage increase in flow from the baseline value. 2.4. Prospective study Of the 757 patients in the cross-sectional study, a subset of 465 patients with normal renal function, defined as having all of the following criteria: serum creatinine levels b1.0 mg/dL, UAE b25 mg/day and eGFR ≥ 60 mL/min/1.73 m2 at the time of the enrollment [1,2,6,20,21], were selected to examine the predictive value of FMD for the progression from normal renal function to the early stage renal dysfunction. All these patients had follow-up visits at our institution every 3 months for a period of 12 months in order to measure serum creatinine and 24 h UAE levels. The endpoint was development of early stage renal dysfunction, defined as occurrence of one or more of the following events: serum creatinine level ≥1.2 mg/dL, UAE level ≥ 30 mg/day and eGFR b 60 mL/min/1.73 m2 [1,2,6,20,21]. CCr was not included in the criteria of development of early stage renal dysfunction during follow-up because of difficulty in blood sampling for serum creatinine measurement during collecting urine for 24 h at home. As shown in Table 1, all the patients continued their standardized medications during the follow-up period. Patients with any of the following events during the
Table 1 Baseline characteristics of the study patients. Normal renal functiona Total patients (n = 448) (n = 757) Age (years) Gender (male) (%) Previous MI (%) Current smoker (%) Hypertension (%) Diabetes mellitus (%) Hyperlipidemia (%) BMI (kg/m2) Systolic BP (mmHg) HbA1c (%) CRP (mg/dL) LVEF (%) Renal functional parameters Serum creatinine (mg/dL) CCr (mL/min/1.73 m2) UAE (mg/day) eGFR (mL/min/1.73 m2) Vascular ultrasound parameters FMD (%) Dilation to NTG (%) Resting arterial diameter (mm) Resting arterial blood flow (mL/min) Increase in arterial blood flow (%) Medication use Statin (%) ACEI/ARB (%) CCB (%) Diuretics (%) Insulin (%)
65.6 ± 9.6 77.0 16.3 27.9 46.4 26.1 49.1 25.4 ± 2.6 131 ± 12 5.9 ± 1.4 0.11 (0.04, 0.30) 61 ± 14
67.3 ± 9.3 74.1 15.9 26.4 49.2 35.4 45.9 25.4 ± 2.6 132 ± 12 6.0 ± 1.4 0.12 (0.05, 0.30) 62 ± 15
0.7 (0.6, 0.8) 89 (77, 103) 6.0 (3.6, 12.1) 85 (71, 110)
0.8 (0.6, 1.3) 78 (57, 94) 9.2 (4.3, 20.5) 66(39, 93)
5.6 ± 1.8 18.7 ± 2.4 4.2 ± 1.4 192 ± 48 240 ± 66
5.1 ± 1.8 18.9 ± 2.9 4.3 ± 1.4 193 ± 47 243 ± 67
30.4 44.4 31.9 8.7 6.5
34.2 46.6 39.1 13.9 7.7
Values are expressed as mean ± SD, median (inter quartile ranges), and percentage of frequencies. Hypertension, defined as N140/90 mmHg or use of antihypertensive medication; diabetes mellitus, defined according to the American Diabetes Association criteria or taking an anti-diabetic medication. Hyperlipidemia, defined according to the National Cholesterol Education Program guidelines or taking lipid lowering medication. MI, myocardial infarction; BP, blood pressure; LVEF, left ventricular ejection fraction; CCr, creatinine clearance; UAE, urinary albumin excretion; eGFR, estimated glomerular filtration rate; FMD, flow-mediated dilation of brachial artery; NTG, nitroglycerin; CCB, calcium channel blocker. a Normal renal function, defined as described in text.
follow-up period were withdrawn from the prospective study; use of contrast media, use of non-steroidal anti-inflammatory drugs for ≥ 3 days, cardiovascular events, major injury or surgery, chronic inflammatory diseases and other serious diseases. Our preliminary data showed 15% of patients with normal renal function at baseline (serum creatinine level b 1.0 mg/dL, UAE b25 mg/day and eGRF ≥ 60 mL/min/1.73 m2) developed serum creatinine levels ≥ 1.2 mg/dL, UAE ≥ 30 mg/day and/or eGRF b60 mL/min/1.73 m2 during the 12 month follow-up period. Based on the data, we calculated 450 patients were required to provide our two-sided multiple logistic models with sufficient statistical power of 0.80 (β = 0.20 and α = 0.05), which justified the number of patients (n = 465) included in this prospective study.
2.5. Statistical analysis Data are expressed as either the mean value ± SD, median and interquartile range (25 and 75th percentiles) or frequencies (%). The Shapiro–Wilk test showed that the levels of serum creatinine, UAE, CCr, eGFR and CRP levels were not distributed normally, and therefore these data were expressed as the median and interquartile range, and they were log-transformed when these data were statistically analyzed. The baseline clinical parameters were compared using Student's unpaired t test or Chi-square analysis where appropriate. In the cross-sectional study, the independent relationship of FMD with serum creatinine levels, UAE, CCr and eGFR at baseline was examined in the multivariate linear regression analysis using the baseline clinical parameters as covariates that had a significant relationship with serum creatinine levels, UAE, CCr and eGFR in the univariate linear regression analysis. In the longitudinal follow-up study, the predictive value of FMD for development of early stage renal dysfunction was examined by multivariate logistic regression analysis using covariates that had a statistically significant difference between patients with and without development of the early stage renal dysfunction. The Hosmer– Lemeshow goodness-of-fit test was used to assess the logistic model fit. Odds ratios (ORs) were estimated with 95% confidence intervals (CIs). The c-statistics using receiver operating characteristic (ROC) curve analysis for logistic model was used to examine the incremental effect of FMD on the predictive value of conventional risk factors for the
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188 development of early stage renal dysfunction. All probability values presented are 2-tailed, with statistical significance being inferred at p b 0.05. All analyses were carried out using STATA version 10.0 (StataCorp, College Station, Texas, USA).
Table 2 Univariate linear regression models of relationships between renal functional parameters and clinical parameters including FMD at baseline in total patients with CAD.
3. Results 3.1. Cross-sectional study for association between FMD and renal parameters at baseline in total patients with CAD Higher serum creatinine levels, raised UAE and lower CCr were significantly associated with impaired FMD, older age, diabetes mellitus, hypertension, and increased CRP levels (Table 2). Also, lower eGFR was significantly associated with the same risk factors except age (Table 2). Age was excluded from statistical analysis for association with eGFR because the calculation of eGFR was based on age. In multivariate linear regression analysis, lower FMD was associated significantly with higher serum creatinine levels, raised UAE, lower CCr and eGFR, with these relationships being independent of diabetes mellitus, hypertension, CRP levels and older age (Table 3). The vasodilator response to nitroglycerin and the increase in brachial blood flow during reactive hyperemia were not associated significantly with serum creatinine levels, UAE, CCr or eGFR (Table 2).
3.2. Longitudinal study in patients with CAD and normal baseline renal function Seventeen patients with normal renal function were withdrawn during the 12 month follow-up period because of the occurrence of cardiovascular events in 8 patients, major surgery in 5 patients and chronic inflammatory diseases in 4 patients. Then, the remaining 448 patients with normal renal function, defined as having all of the following criteria: serum creatinine levels b1.0 mg/dL, UAE b25 mg/ day and eGFR ≥60 mL/min/1.73 m2 at the time of the enrollment, were included for further analyses of the prospective study. On the basis of eGFR, 46% and 54% of them were classified to CKD stages 1 and 2, respectively. During this follow-up period, 96 patients developed progression from normal renal function to early stage renal dysfunction, defined as occurrence of one or more of the following events: serum creatinine level ≥1.2 mg/dL, UAE level ≥30 mg/day and/or eGFR b60 mL/min/1.73 m2. Among these 96 patients, the increase in levels of creatinine and UAE and the decrease in eGFR occurred in 67, 48 and in 93 patients, respectively. All of them had an increase in CKD stage during the 12 month follow-up period, and 3% and 97% of them were classified to CKD stages 2 and 3, respectively, at the end of the follow-up period. However, progression to serum creatinine levels ≥2.0 mg/dL, UAE ≥300 mg/day or eGFR b30 mL/ min/1.73 m2 was not observed in any of these patients during the follow-up period. We observed that patients who progressed to early stage renal dysfunction had lower FMD, a higher prevalence of hypertension and diabetes mellitus, increased CRP levels and were older than patients who did not develop the early stage renal dysfunction (Table 4). There were no differences in the frequency of use of medications including angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, statin and diuretics between patients with and without progression to the early stage renal dysfunction (Table 4). Moreover, multivariate logistic regression analysis showed that lower FMD was associated significantly with progression to the early stage renal dysfunction after adjustment of the data for age, hypertension, diabetes mellitus, and CRP levels as covariates (Fig. 1). These covariates were included because they had a significant difference between patients with and without development of the early stage renal dysfunction (Table 4). Using a C-statistic analysis, we showed that impaired FMD levels had a significant incremental effect on the predictive value of the conventional known risks including age,
185
FMD Age (years) Diabetes mellitus Hypertension CRP LVEF Hyperlipidemia Current smoker Use of ACEI/ARB Use of diuretics Brachial arterial parameters Dilation to NTG Resting arterial diameter Resting arterial blood flow Increase in arterial blood flow
Serum creatinine
UAE
CCr
eGFR
r
r
r
r
− 0.36a 0.16a 0.23a 0.22a 0.20a 0.08 0.04 − 0.04 0.05 0.05
− 0.35a 0.10b 0.17a 0.17a 0.14a 0.02 0.02 − 0.05 0.06 0.01
0.43a − 0.13a − 0.22a − 0.17a − 0.19a − 0.07 0.01 0.03 −0.02 − 0.04
0.34a – − 0.19a − 0.22a − 0.12a − 0.07 0.03 0.05 − 0.09 − 0.01
− 0.04 − 0.03 0.02 0.01
− 0.04 0.02 0.04 0.02
0.05 0.03 − 0.08 −0.02
0.03 0.02 − 0.02 − 0.01
Data are expressed as regression coefficient (r). The levels of serum creatinine, UAE, CCr, eGFR and CRP levels were log-transformed for statistical analysis as these variables were not normally distributed. Abbreviations as in Table 1. a p b 0.01. b p b 0.05.
hypertension, diabetes mellitus, and CRP levels for progression of renal dysfunction (Table 5 and Fig. 2).
4. Discussion The present cross-sectional and longitudinal studies demonstrated that impaired FMD in the brachial artery was associated with early stage renal dysfunction and was a predictor of development from normal renal function to early stage renal dysfunction within the 12 month follow-up period in patients with CAD. Furthermore, the predictive value of impaired FMD was incremental over that of the previously known conventional predictors for renal dysfunction. Renal dysfunction is known to increase oxidative stress, that in turn, may cause endothelial vasomotor dysfunction in the systemic vasculature [22,23]. This positive feedback mechanism in either the renal dysfunction or endothelial dysfunction in systemic arteries could therefore amplify the vicious circle that exists between renal and endothelial dysfunctions. However, it was unclear whether endothelial dysfunction in the systemic vasculature was a cause or a consequence of renal dysfunction. In this context, the present study is
Table 3 Multiple linear regression models of relationships between renal functional parameters and clinical parameters including FMD at baseline in total patients with CAD.
FMD Diabetes mellitus Hypertension CRP Age (years)
Serum creatinine
UAE
CCr
eGFR
β
β
β
β
− 0.29a 0.16a 0.18a 0.11c 0.11b
− 0.30a 0.14b 0.13a 0.11b 0.08
0.39a − 0.13a − 0.11b − 0.11b − 0.10b
0.30a − 0.18a − 0.19b − 0.10c –
Data are expressed as standardized regression coefficient (β). Clinical parameters were selected as independent variables because these parameters had a significant relation with renal parameters in univariate linear regression analysis in Table 2. The levels of creatinine, UAE, CCr, eGFR and CRP were log-transformed for statistical analysis as these variables were not normally distributed. Abbreviations as in Table 1. a p b 0.001. b p b 0.01. c p b 0.05.
186
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
Table 4 Comparison of baseline clinical characteristics of patients with and without development of early renal dysfunction during 12 months follow-up in patients with CAD and normal baseline renal function.
FMD (%) Hypertension (%) Diabetes mellitus (%) CRP (mg/dL) Age (years) LVEF (%) Current smoker (%) Hyperlipidemia (%) Use of ACEI/ARB (%) Use of statin (%) Use of diuretics (%) Dilation to NTG (%)
Serum creatinine (mg/dL)
UAE (mg/day)
≥1.2 (n = 67)
≥30 (n = 48)
a
4.2 ± 1.7 64.2b 44.8a 0.3 (0.1, 0.5)a 69.0 ± 7.0b 60 ± 10 29.9 49.3 49.3 28.9 16.4 18.1 ± 1.4
b 1.2 (n = 381) 5.8 ± 1.7 43.3 22.8 0.1 (0.1, 0.3) 65.3 ± 9.1 61 ± 11 27.6 48.6 43.6 31.2 10.5 19.1 ± 1.4
a
4.0 ± 1.3 72.9a 56.3a 0.3 (0.2, 0.6)b 70.1 ± 5.3a 60 ± 11 20.8 56.3 43.8 26.0 14.6 18.5 ± 1.5
eGFR (mL/min/1.73 m2) b30 (n = 400) 5.8 ± 1.8 43.3 22.5 0.1 (0.1, 0.3) 65.4 ± 9.1 62 ± 10 28.8 47.8 40.0 30.0 11.0 18.9 ± 1.3
b 60 (n = 93) a
4.3 ± 1.6 55.9a 63.4a 0.3 (0.1, 0.6)b 69.3 ± 7.9a 60 ± 10 24.8 50.5 46.2 30.1 16.1 18.0 ± 1.9
≥ 60 (n = 355) 5.9 ± 1.8 43.9 23.1 0.1 (0.1, 0.3) 65.3 ± 9.0 62 ± 11 20.0 48.2 38.9 30.7 10.1 18.9 ± 2.0
Values are expressed as mean ± SD, median (inter quartile ranges), and percentage of frequencies. This analysis included a subgroup of 448 patients with normal renal function, defined as described in text and Table 1. Other abbreviations as in Table 1. a p b 0.001, vs. patients with serum creatinine levels b 1.2 mg/dL, UAE b 30 mg/day and eGFR ≥60 mL/min/1.73 m2, respectively. b p b 0.01.
the first to show that endothelial vasomotor dysfunction in systemic arteries may precede the development of renal dysfunction in patients with normal renal function.
The underlying mechanism to explain how endothelial dysfunction in systemic arteries causes renal dysfunction remains unclear. In addition to its vasodilatory action, endothelium-derived nitric oxide
Fig. 1. Odds ratio (OR) and 95% confidence intervals (95% CI) predicting progression from normal renal function to early stage renal dysfunction during 12 months follow-up in 448 patients with CAD and normal renal function at baseline, calculated by multivariate logistic regression analysis using flow-mediated dilation (FMD) in the brachial artery, age, diabetes, hypertension and C-reactive protein (CRP) levels as the independent variables. These independent variables were included because they had a significant difference between patients with and without development of the early stage renal dysfunction (Table 4). (A) progression to serum creatinine levels ≥ 1.2 mg/dL, (B) progression to urine albumin excretion (UAE) ≥30 mg/day, and (C) progression to eGFR b 60 mL/min/1.73 m2. ⁎p b 0.05. †p b 0.01.
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
also prevents inflammation, cell proliferation and fibrosis, processes that are closely related to glomerular, arteriolar and tubulointerstitial renal damage [9]. Moreover, evidence from animal experiments shows that systemic administration of a nitric oxide synthase inhibitor induces renal vasoconstriction and injury that is characterized by glomerulosclerosis, interstitial fibrosis, arterial and arteriolar lesions and cellular infiltration in the tubulointerstitial areas of the kidney [24,25]. Endothelial vasomotor dysfunction in the brachial artery may therefore reflect a decrease in nitric oxide activity in the renal vasculature that leads to renal dysfunction. These results indicate that endothelial dysfunction in systemic arteries predisposes to the development of renal dysfunction. And, these findings suggest endothelial dysfunction in systemic arteries is a common denominator of the early stages of atherosclerotic vascular dysfunction and renal dysfunction, both of which are strong risk factors for future cardiovascular events [1,2,26]. In a cross-sectional study of hypertensive patients, Perticone et al. [16] showed a significant association between the decrease in eGFR and impairment in endothelium-dependent dilator responses in microvessels of the forearm. These findings are in agreement with the results of our cross-sectional analyses and suggest a strong association exists between renal dysfunction and endothelial vasomotor dysfunction, irrespective of vessel size. Our study had some limitations. The present study arbitrarily chose cut-off levels of serum creatinine levels and UAE for criteria of normal or decreased renal function according to previous reports [1,2,6,20,21]. On the basis of eGFR [1], most of the present patients meeting the criteria of early stage renal dysfunction had an increase in CKD stage from stage 1 or 2 to stage 3 during the follow-up period. Therefore, cut-off levels of serum creatinine levels and UAE used in the present study seemed to be acceptable. It remains undetermined whether endothelial dysfunction in the brachial artery may be related to future end-stage renal disease for longer time periods. It also remains unclear whether endothelial dysfunction in the brachial artery is related to early renal dysfunction in the general population. A large prospective trial incorporating medications which improve endothelial function is therefore required in order to more precisely assess the role of endothelial dysfunction in systemic arteries in the pathogenesis of renal dysfunction progression. In conclusion, endothelial vasomotor dysfunction in the brachial artery is independently associated with renal dysfunction, and also is an independent predictor for the development of early renal dysfunction in patients with CAD. Measurement of FMD is therefore a useful marker for risk stratification of future renal dysfunction.
187
15012222” from the Ministry of Education, Culture, Sports, Science, and Technology, Health and Labor Sciences Research Grants for Comprehensive Research on Aging and Health (H15-Choju-012), Tokyo, Japan. The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [27].
Acknowledgement This study was supported by grants-in-aid for (B)(2)-15390244 and (B)-19390209, Priority Areas (C) “Medical Genome Science
Table 5 C-statistics for logistic models predicting development of early renal dysfunction during 12 month follow-up using conventional risk factors with or without FMD in patients with CAD and normal baseline renal function.
Creatinine ≥1.2 mg/dL UAE ≥30 mg/day eGFR b60 mL/min/1.73 m2
Conventional risks without FMD
Conventional risks with FMD
p-valuea
0.77 0.85 0.70
0.82 0.90 0.79
0.03 0.01 0.01
Data mean area under the ROC curve (AUC). Conventional risks included age, hypertension, diabetes mellitus, and CRP that had a significant difference between patients with and without development of early renal dysfunction, as shown in Table 4. Abbreviations as in Table 1. a p-value for the differences in AUC between conventional risks with and without FMD.
Fig. 2. Comparison of ROC curves for progression from normal renal function to early stage renal dysfunction during 12 months follow-up in 448 patients with CAD and normal renal function at baseline. (A) ROC curves for progression to serum creatinine levels ≥1.2 mg/dL. (B) ROC curves for progression to UAE ≥30 mg/day. (C) ROC curves for progression to eGFR b60 mL/min/1.73 m2. The curves are based on logistic models using covariables including age, diabetes, hypertension, and C-reactive protein (CRP) levels with or without FMD. Solid lines (blue) indicate covariates including FMD levels, and dashed lines (red) indicate the covariates without FMD levels. AUC, area under the ROC curve.
188
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
References [1] Gupta R, Birnbaum Y, Uretsky BF. The renal patients with coronary artery disease. Current concepts and dilemmas. J Am Coll Cardiol 2004;44:1343–53. [2] Schiffrin EL, Lipman ML, Mann JF. Chronic kidney disease: effects on the cardiovascular system. Circulation 2007;116:85–97. [3] Jafar TH, Stark PC, Schmid CH, et al. AIPRD Study Group. Progression of chronic kidney disease: the role of blood pressure control, proteinuria, and angiotensinconverting enzyme inhibition: a patient-level meta-analysis. Ann Intern Med 2003;139:244–52. [4] Mann JF, Schmieder RE, McQueen M, et al. ONTARGET investigators. Renal outcomes with telmisartan, ramipril, or both, in people at high vascular risk (the ONTARGET study): a multicentre, randomised, double-blind, controlled trial. Lancet 2008;372:547–53. [5] Sukhija R, Bursac Z, Kakar P, et al. Effect of statins on the development of renal dysfunction. Am J Cardiol 2008;101:975–9. [6] Levey AS, Eckardt K, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from kidney disease: improving global outcomes (KDIGO). Kidney Int 2005;67:2089–100. [7] Taal MW, Brenner BM. Predicting initiation and progression of chronic kidney disease: developing renal risk scores. Kidney Int 2006;70:1694–705. [8] Deckert T, Feldt-Rasmussen B, Borch-Johnsen K, Jensen T, Kofoed-Enevoldsen A. Albuminuria reflects widespread vascular damage. The Steno hypothesis. Diabetologia 1999;32:219–26. [9] Ochodnický P, de Zeeuw D, Henning RH, Kluppel CA, van Dokkum RP. Endothelial function predicts the development of renal damage after combined nephrectomy and myocardial infarction. J Am Soc Nephrol 2006;17:S49–52. [10] King AJ, Brenner BM. Endothelium-derived vasoactive factors and the renal vasculature. Am J Physiol 1991;29:R653–62. [11] Kang DH, Kanellis J, Hugo C, et al. Role of the microvascular endothelium in progressive renal disease. J Am Soc Nephrol 2003;13:806–16. [12] Anderson TJ, Uehata A, Gerhard MD, et al. Close relation of endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol 1995;26:1235–41. [13] Stam F, van Guldener C, Becker A, et al. Endothelial dysfunction contributes to renal function-associated cardiovascular mortality in a population with mild renal insufficiency: the Hoorn study. J Am Soc Nephrol 2006;17:537–45.
[14] Ochodnicky P, Vettoretti S, Henning RH, Buikema H, Van Dokkum RP, de Zeeuw D. Endothelial dysfunction in chronic kidney disease: determinant of susceptibility to end-organ damage and therapeutic response. J Nephrol 2006;19:246–58. [15] Zoccali C. The endothelium as a target in renal diseases. J Nephrol 2007;20(suppl 12): S39–44. [16] Perticone F, Maio R, Tripepi G, Zoccali C. Endothelial dysfunction and mild renal insufficiency in essential hypertension. Circulation 2004;110:821–5. [17] van der Harst P, Smilde TD, Buikema H, et al. Vascular function and mild renal impairment in stable coronary artery disease. Arterioscler Thromb Vasc Biol 2006;26:379–84. [18] Kugiyama K, Motoyama T, Doi H, et al. Improvement of endothelial vasomotor dysfunction by treatment with alpha-tocopherol in patients with high remnant lipoproteins levels. J Am Coll Cardiol 1999;33:1512–8. [19] Nakamura T, Takano H, Umetani K, et al. Remnant lipoproteinemia is a risk factor for endothelial vasomotor dysfunction and coronary artery disease in metabolic syndrome. Atherosclerosis 2005;181:321–7. [20] Couchoud C, Pozet N, Labeeuw M, Pouteil-Noble C. Screening early renal failure: cut-off values for serum creatinine as an indicator of renal impairment. Kidney Int 1999;55:1878–84. [21] Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 1992;38:1933–53. [22] Cottone S, Lorito MC, Riccobene R, et al. Oxidative stress, inflammation and cardiovascular disease in chronic renal failure. J Nephrol 2008;21:175–9. [23] Oberg BP, McMenamin E, Lucas FL, et al. Increased prevalence of oxidant stress and inflammation in patients with moderate to severe chronic kidney disease. Kidney Int 2004;65:1009–16. [24] Jover B, Mimran A. Nitric oxide inhibition and renal alterations. J Cardiovasc Pharmacol 2001;38:S65–70. [25] Fujihara CK, Michellazzo SM, De-Nucci G, Zatz R. Sodium excess aggravates hypertension and renal parenchymal injury in rats with chronic NO inhibition. Am J Physiol 1994;266:F697–705. [26] Brunner H, Cockcroft JR, Deanfield J, et al. Endothelial function and dysfunction. Part II: association with cardiovascular risk factors and diseases. J Hypertens 2005;23:233–46. [27] Coats AJ. Ethical authorship and publishing. Int J Cardiol 2009;131:149–50.
Contents lists available at ScienceDirect
International Journal of Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j c a r d
Endothelial vasomotor dysfunction in the brachial artery predicts the short-term development of early stage renal dysfunction in patients with coronary artery disease Takamitsu Nakamura, Jun-ei Obata, Mitsumasa Hirano, Yoshinobu Kitta, Keita Sano, Tsuyoshi Kobayashi, Daisuke Fujioka, Yukio Saito, Toshiaki Yano, Kenichi Kawabata, Kazuhiro Watanabe, Yosuke Watanabe, Hideto Mishina, Kiyotaka Kugiyama ⁎ University of Yamanashi, Faculty of Medicine, Department of Internal Medicine II, Chuo, Yamanashi, Japan
a r t i c l e
i n f o
Article history: Received 23 July 2009 Accepted 31 October 2009 Available online 27 November 2009 Keywords: Endothelial vasomotor function Renal failure Predictor Coronary artery disease
a b s t r a c t Background: This study examined whether endothelial vasomotor dysfunction in the brachial artery predicted early renal dysfunction in patients with coronary artery disease (CAD). Endothelial function in the renal vasculature plays an important role in the regulation of renal hemodynamics. As endothelial dysfunction is a systemic disorder, there may be a relationship between endothelial function in the brachial artery and renal vasculature. Methods: Flow-mediated endothelium-dependent dilation (FMD) in brachial artery and renal functional parameters were measured in 757 patients with CAD without macroalbuminuria. Results: In a cross-sectional data, an impaired FMD was associated with higher serum creatinine levels and urinary albumin excretion (UAE), lower creatinine clearance rate and estimated glomerular filtration rate (eGFR) at baseline in multiple linear regression analysis. In a follow-up study including a subgroup of 448 patients with normal renal function (serum creatinine level b1.0 mg/dL, UAE b 25 mg/day and eGFR ≥ 60 mL/ min/1.73 m2 at baseline), 96 patients had an endpoint of early stage renal dysfunction (serum creatinine levels ≥ 1.2 mg/dL, UAE ≥ 30 mg/day and/or eGFR b 60 mL/min/1.73 m2) during 12 month follow-up. Multivariate logistic regression analysis showed that impaired FMD was significantly associated with progression to the early stage renal dysfunction after adjustment with age, diabetes mellitus, hypertension and C-reactive protein levels. Conclusions: Endothelial vasomotor dysfunction in the brachial artery is independently associated with progression from normal renal function to early stage renal dysfunction in patients with CAD. Measurement of FMD may therefore be useful for assessing risk of future renal dysfunction. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Chronic kidney disease is associated with a high prevalence of cardiovascular events [1,2]. There is evidence that treatment at the earliest possible stage of renal dysfunction is more effective for preventing progression to end-stage renal disease and subsequent cardiovascular disease [3–7]. It is therefore clinically important to detect this group of patients who are at high risk of developing early stage renal disease [7]. Microalbuminuria, a marker of early stage renal dysfunction, is related to renal microvasculopathy, characterized by preglomerular arteriolar involvement and tubulointerstitial
⁎ Corresponding author. Department of Internal Medicine II, University of Yamanashi, Faculty of Medicine, 1110 Shimokato, Chuo 409-3898, Japan. Tel.: + 81 55 273 9590; fax: + 81 55 273 6749. E-mail address: [email protected] (K. Kugiyama). 0167-5273/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2009.10.054
changes [1,2,8]. Experimental studies indicate that endothelial dysfunction in the renal vasculature is involved in the development of renal microvasculopathies and microalbuminuria [9–11]. As endothelial dysfunction is a systemic disorder [12], it is possible that there is a relationship between endothelial dysfunction in the systemic arteries and renal vasculature [13–15]. A previous crosssectional study showed that an impaired endothelial vasomotor response in the forearm was associated with mild to moderate renal dysfunction in hypertensive patients without previous cardiovascular diseases [16]. In contrast, in vitro measurement of endotheliumdependent dilation in isolated internal thoracic arteries showed no significant relationship with the presence of mild to moderate renal dysfunction in patients with coronary artery disease (CAD) [17]. Therefore, it remains controversial whether there is an association between endothelial vasomotor dysfunction in systemic arteries and renal dysfunction in patients with cardiovascular disease and advanced atherosclerotic burden. It also remains unknown whether
184
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
endothelial vasomotor dysfunction in systemic arteries precedes renal dysfunction, thereby providing a predictive value of future development of renal dysfunction. In the present study, we examined whether endothelial vasomotor dysfunction in the brachial artery provided prognostic information on the development of early stage renal dysfunction in patients with CAD that is linked with chronic kidney disease. 2. Methods 2.1. Study patients This study included 757 consecutively enrolled patients with stable CAD without macroalbuminuria (≥300 mg/day) at Yamanashi University Hospital from January 1, 2002 to Dec 31, 2006. All the patients had a routine screen following enrollment in the study that included measurement of flow-dependent dilation (FMD) in the brachial artery using Bmode ultrasound images, serum creatinine levels, 24 h urine albumin excretion (UAE) and calculation of creatinine clearance (CCr) and estimated glomerular filtration rate (eGFR). All patients had angiographic evidence of organic diameter stenosis N 70% of at least one major coronary artery (single-vessel disease n = 175; two-vessel disease n = 302; and three-vessel disease n = 280). The exclusion criteria were: 1) a history of dialysis, 2) use of contrast media 3 months prior to enrollment, 3) congestive heart failure (≥New York Heart Association classification III), 4) secondary hypertension, 5) renal artery stenosis, 6) polycystic kidney disease, 7) chronic inflammatory diseases, 8) major injury or surgery 3 months prior to enrollment, and 9) other serious diseases. The baseline characteristics of the patients in the study are summarized in Table 1. All the patients gave written, informed consent for the study at enrollment and received the standardized medications outlined in Table 1. This study was conducted in accordance with the guidelines approved by the ethics committee of Yamanashi University Hospital and conformed with the principles outlined in the 1975 Declaration of Helsinki. 2.2. Measurements of renal function Creatinine levels in serum and urine were measured by an enzymatic method using an auto-analyzer (Mitsubishi Kagaku Iatron, Tokyo, Japan). Serum creatinine levels were measured on 2 days a week apart, with the average value being used in the analyses. CCr was calculated from 24 h urine creatinine concentration and the serum creatinine concentration on the same day, and the value normalized to a surface area of 1.73 m2 [6]. GFR was estimated by Modification of Diet in Renal Disease study equation [1]. Urinary albumin levels were measured by immunonephelometry, and fasting plasma high sensitivity C-reactive protein (CRP) levels by rate nephelometry (Dade Behring, Marburg, Germany). 2.3. Measurements of endothelial vasomotor function of the brachial artery assessed by flow-mediated dilation (FMD) Vasodilator responses in the brachial arteries were measured using B-mode ultrasound images with a 7.5-MHz linear array transducer (HP-5500, Phillips Corp., Tokyo, Japan) according to a method validated in our previous studies [18,19]. Briefly, after baseline measurements of the diameter and flow velocity in the brachial artery, a blood pressure cuff was placed around the forearm and inflated to a pressure of 250 to 300 mm Hg for 5 min and then released. Measurements of the artery diameter during the reactive hyperemia were taken 45 to 90 s after cuff deflation. Sublingual nitroglycerin (300 μg) was then administered and the measurements repeated 3 min later. The responses of the vessel diameters to reactive hyperemia and nitroglycerin were expressed as the percentage increase in diameter from the baseline value. The diameters of the vessel responses were assessed at three points along a 10-mm length of the artery, and the values averaged. Blood flow was calculated by multiplying the velocity–time integral of the Doppler flow signal by heart rate and cross-sectional area of the vessel. The increase in brachial blood flow was calculated as the maximum flow recorded in the first 15 s after cuff deflation and was expressed as a percentage increase in flow from the baseline value. 2.4. Prospective study Of the 757 patients in the cross-sectional study, a subset of 465 patients with normal renal function, defined as having all of the following criteria: serum creatinine levels b1.0 mg/dL, UAE b25 mg/day and eGFR ≥ 60 mL/min/1.73 m2 at the time of the enrollment [1,2,6,20,21], were selected to examine the predictive value of FMD for the progression from normal renal function to the early stage renal dysfunction. All these patients had follow-up visits at our institution every 3 months for a period of 12 months in order to measure serum creatinine and 24 h UAE levels. The endpoint was development of early stage renal dysfunction, defined as occurrence of one or more of the following events: serum creatinine level ≥1.2 mg/dL, UAE level ≥ 30 mg/day and eGFR b 60 mL/min/1.73 m2 [1,2,6,20,21]. CCr was not included in the criteria of development of early stage renal dysfunction during follow-up because of difficulty in blood sampling for serum creatinine measurement during collecting urine for 24 h at home. As shown in Table 1, all the patients continued their standardized medications during the follow-up period. Patients with any of the following events during the
Table 1 Baseline characteristics of the study patients. Normal renal functiona Total patients (n = 448) (n = 757) Age (years) Gender (male) (%) Previous MI (%) Current smoker (%) Hypertension (%) Diabetes mellitus (%) Hyperlipidemia (%) BMI (kg/m2) Systolic BP (mmHg) HbA1c (%) CRP (mg/dL) LVEF (%) Renal functional parameters Serum creatinine (mg/dL) CCr (mL/min/1.73 m2) UAE (mg/day) eGFR (mL/min/1.73 m2) Vascular ultrasound parameters FMD (%) Dilation to NTG (%) Resting arterial diameter (mm) Resting arterial blood flow (mL/min) Increase in arterial blood flow (%) Medication use Statin (%) ACEI/ARB (%) CCB (%) Diuretics (%) Insulin (%)
65.6 ± 9.6 77.0 16.3 27.9 46.4 26.1 49.1 25.4 ± 2.6 131 ± 12 5.9 ± 1.4 0.11 (0.04, 0.30) 61 ± 14
67.3 ± 9.3 74.1 15.9 26.4 49.2 35.4 45.9 25.4 ± 2.6 132 ± 12 6.0 ± 1.4 0.12 (0.05, 0.30) 62 ± 15
0.7 (0.6, 0.8) 89 (77, 103) 6.0 (3.6, 12.1) 85 (71, 110)
0.8 (0.6, 1.3) 78 (57, 94) 9.2 (4.3, 20.5) 66(39, 93)
5.6 ± 1.8 18.7 ± 2.4 4.2 ± 1.4 192 ± 48 240 ± 66
5.1 ± 1.8 18.9 ± 2.9 4.3 ± 1.4 193 ± 47 243 ± 67
30.4 44.4 31.9 8.7 6.5
34.2 46.6 39.1 13.9 7.7
Values are expressed as mean ± SD, median (inter quartile ranges), and percentage of frequencies. Hypertension, defined as N140/90 mmHg or use of antihypertensive medication; diabetes mellitus, defined according to the American Diabetes Association criteria or taking an anti-diabetic medication. Hyperlipidemia, defined according to the National Cholesterol Education Program guidelines or taking lipid lowering medication. MI, myocardial infarction; BP, blood pressure; LVEF, left ventricular ejection fraction; CCr, creatinine clearance; UAE, urinary albumin excretion; eGFR, estimated glomerular filtration rate; FMD, flow-mediated dilation of brachial artery; NTG, nitroglycerin; CCB, calcium channel blocker. a Normal renal function, defined as described in text.
follow-up period were withdrawn from the prospective study; use of contrast media, use of non-steroidal anti-inflammatory drugs for ≥ 3 days, cardiovascular events, major injury or surgery, chronic inflammatory diseases and other serious diseases. Our preliminary data showed 15% of patients with normal renal function at baseline (serum creatinine level b 1.0 mg/dL, UAE b25 mg/day and eGRF ≥ 60 mL/min/1.73 m2) developed serum creatinine levels ≥ 1.2 mg/dL, UAE ≥ 30 mg/day and/or eGRF b60 mL/min/1.73 m2 during the 12 month follow-up period. Based on the data, we calculated 450 patients were required to provide our two-sided multiple logistic models with sufficient statistical power of 0.80 (β = 0.20 and α = 0.05), which justified the number of patients (n = 465) included in this prospective study.
2.5. Statistical analysis Data are expressed as either the mean value ± SD, median and interquartile range (25 and 75th percentiles) or frequencies (%). The Shapiro–Wilk test showed that the levels of serum creatinine, UAE, CCr, eGFR and CRP levels were not distributed normally, and therefore these data were expressed as the median and interquartile range, and they were log-transformed when these data were statistically analyzed. The baseline clinical parameters were compared using Student's unpaired t test or Chi-square analysis where appropriate. In the cross-sectional study, the independent relationship of FMD with serum creatinine levels, UAE, CCr and eGFR at baseline was examined in the multivariate linear regression analysis using the baseline clinical parameters as covariates that had a significant relationship with serum creatinine levels, UAE, CCr and eGFR in the univariate linear regression analysis. In the longitudinal follow-up study, the predictive value of FMD for development of early stage renal dysfunction was examined by multivariate logistic regression analysis using covariates that had a statistically significant difference between patients with and without development of the early stage renal dysfunction. The Hosmer– Lemeshow goodness-of-fit test was used to assess the logistic model fit. Odds ratios (ORs) were estimated with 95% confidence intervals (CIs). The c-statistics using receiver operating characteristic (ROC) curve analysis for logistic model was used to examine the incremental effect of FMD on the predictive value of conventional risk factors for the
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188 development of early stage renal dysfunction. All probability values presented are 2-tailed, with statistical significance being inferred at p b 0.05. All analyses were carried out using STATA version 10.0 (StataCorp, College Station, Texas, USA).
Table 2 Univariate linear regression models of relationships between renal functional parameters and clinical parameters including FMD at baseline in total patients with CAD.
3. Results 3.1. Cross-sectional study for association between FMD and renal parameters at baseline in total patients with CAD Higher serum creatinine levels, raised UAE and lower CCr were significantly associated with impaired FMD, older age, diabetes mellitus, hypertension, and increased CRP levels (Table 2). Also, lower eGFR was significantly associated with the same risk factors except age (Table 2). Age was excluded from statistical analysis for association with eGFR because the calculation of eGFR was based on age. In multivariate linear regression analysis, lower FMD was associated significantly with higher serum creatinine levels, raised UAE, lower CCr and eGFR, with these relationships being independent of diabetes mellitus, hypertension, CRP levels and older age (Table 3). The vasodilator response to nitroglycerin and the increase in brachial blood flow during reactive hyperemia were not associated significantly with serum creatinine levels, UAE, CCr or eGFR (Table 2).
3.2. Longitudinal study in patients with CAD and normal baseline renal function Seventeen patients with normal renal function were withdrawn during the 12 month follow-up period because of the occurrence of cardiovascular events in 8 patients, major surgery in 5 patients and chronic inflammatory diseases in 4 patients. Then, the remaining 448 patients with normal renal function, defined as having all of the following criteria: serum creatinine levels b1.0 mg/dL, UAE b25 mg/ day and eGFR ≥60 mL/min/1.73 m2 at the time of the enrollment, were included for further analyses of the prospective study. On the basis of eGFR, 46% and 54% of them were classified to CKD stages 1 and 2, respectively. During this follow-up period, 96 patients developed progression from normal renal function to early stage renal dysfunction, defined as occurrence of one or more of the following events: serum creatinine level ≥1.2 mg/dL, UAE level ≥30 mg/day and/or eGFR b60 mL/min/1.73 m2. Among these 96 patients, the increase in levels of creatinine and UAE and the decrease in eGFR occurred in 67, 48 and in 93 patients, respectively. All of them had an increase in CKD stage during the 12 month follow-up period, and 3% and 97% of them were classified to CKD stages 2 and 3, respectively, at the end of the follow-up period. However, progression to serum creatinine levels ≥2.0 mg/dL, UAE ≥300 mg/day or eGFR b30 mL/ min/1.73 m2 was not observed in any of these patients during the follow-up period. We observed that patients who progressed to early stage renal dysfunction had lower FMD, a higher prevalence of hypertension and diabetes mellitus, increased CRP levels and were older than patients who did not develop the early stage renal dysfunction (Table 4). There were no differences in the frequency of use of medications including angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, statin and diuretics between patients with and without progression to the early stage renal dysfunction (Table 4). Moreover, multivariate logistic regression analysis showed that lower FMD was associated significantly with progression to the early stage renal dysfunction after adjustment of the data for age, hypertension, diabetes mellitus, and CRP levels as covariates (Fig. 1). These covariates were included because they had a significant difference between patients with and without development of the early stage renal dysfunction (Table 4). Using a C-statistic analysis, we showed that impaired FMD levels had a significant incremental effect on the predictive value of the conventional known risks including age,
185
FMD Age (years) Diabetes mellitus Hypertension CRP LVEF Hyperlipidemia Current smoker Use of ACEI/ARB Use of diuretics Brachial arterial parameters Dilation to NTG Resting arterial diameter Resting arterial blood flow Increase in arterial blood flow
Serum creatinine
UAE
CCr
eGFR
r
r
r
r
− 0.36a 0.16a 0.23a 0.22a 0.20a 0.08 0.04 − 0.04 0.05 0.05
− 0.35a 0.10b 0.17a 0.17a 0.14a 0.02 0.02 − 0.05 0.06 0.01
0.43a − 0.13a − 0.22a − 0.17a − 0.19a − 0.07 0.01 0.03 −0.02 − 0.04
0.34a – − 0.19a − 0.22a − 0.12a − 0.07 0.03 0.05 − 0.09 − 0.01
− 0.04 − 0.03 0.02 0.01
− 0.04 0.02 0.04 0.02
0.05 0.03 − 0.08 −0.02
0.03 0.02 − 0.02 − 0.01
Data are expressed as regression coefficient (r). The levels of serum creatinine, UAE, CCr, eGFR and CRP levels were log-transformed for statistical analysis as these variables were not normally distributed. Abbreviations as in Table 1. a p b 0.01. b p b 0.05.
hypertension, diabetes mellitus, and CRP levels for progression of renal dysfunction (Table 5 and Fig. 2).
4. Discussion The present cross-sectional and longitudinal studies demonstrated that impaired FMD in the brachial artery was associated with early stage renal dysfunction and was a predictor of development from normal renal function to early stage renal dysfunction within the 12 month follow-up period in patients with CAD. Furthermore, the predictive value of impaired FMD was incremental over that of the previously known conventional predictors for renal dysfunction. Renal dysfunction is known to increase oxidative stress, that in turn, may cause endothelial vasomotor dysfunction in the systemic vasculature [22,23]. This positive feedback mechanism in either the renal dysfunction or endothelial dysfunction in systemic arteries could therefore amplify the vicious circle that exists between renal and endothelial dysfunctions. However, it was unclear whether endothelial dysfunction in the systemic vasculature was a cause or a consequence of renal dysfunction. In this context, the present study is
Table 3 Multiple linear regression models of relationships between renal functional parameters and clinical parameters including FMD at baseline in total patients with CAD.
FMD Diabetes mellitus Hypertension CRP Age (years)
Serum creatinine
UAE
CCr
eGFR
β
β
β
β
− 0.29a 0.16a 0.18a 0.11c 0.11b
− 0.30a 0.14b 0.13a 0.11b 0.08
0.39a − 0.13a − 0.11b − 0.11b − 0.10b
0.30a − 0.18a − 0.19b − 0.10c –
Data are expressed as standardized regression coefficient (β). Clinical parameters were selected as independent variables because these parameters had a significant relation with renal parameters in univariate linear regression analysis in Table 2. The levels of creatinine, UAE, CCr, eGFR and CRP were log-transformed for statistical analysis as these variables were not normally distributed. Abbreviations as in Table 1. a p b 0.001. b p b 0.01. c p b 0.05.
186
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
Table 4 Comparison of baseline clinical characteristics of patients with and without development of early renal dysfunction during 12 months follow-up in patients with CAD and normal baseline renal function.
FMD (%) Hypertension (%) Diabetes mellitus (%) CRP (mg/dL) Age (years) LVEF (%) Current smoker (%) Hyperlipidemia (%) Use of ACEI/ARB (%) Use of statin (%) Use of diuretics (%) Dilation to NTG (%)
Serum creatinine (mg/dL)
UAE (mg/day)
≥1.2 (n = 67)
≥30 (n = 48)
a
4.2 ± 1.7 64.2b 44.8a 0.3 (0.1, 0.5)a 69.0 ± 7.0b 60 ± 10 29.9 49.3 49.3 28.9 16.4 18.1 ± 1.4
b 1.2 (n = 381) 5.8 ± 1.7 43.3 22.8 0.1 (0.1, 0.3) 65.3 ± 9.1 61 ± 11 27.6 48.6 43.6 31.2 10.5 19.1 ± 1.4
a
4.0 ± 1.3 72.9a 56.3a 0.3 (0.2, 0.6)b 70.1 ± 5.3a 60 ± 11 20.8 56.3 43.8 26.0 14.6 18.5 ± 1.5
eGFR (mL/min/1.73 m2) b30 (n = 400) 5.8 ± 1.8 43.3 22.5 0.1 (0.1, 0.3) 65.4 ± 9.1 62 ± 10 28.8 47.8 40.0 30.0 11.0 18.9 ± 1.3
b 60 (n = 93) a
4.3 ± 1.6 55.9a 63.4a 0.3 (0.1, 0.6)b 69.3 ± 7.9a 60 ± 10 24.8 50.5 46.2 30.1 16.1 18.0 ± 1.9
≥ 60 (n = 355) 5.9 ± 1.8 43.9 23.1 0.1 (0.1, 0.3) 65.3 ± 9.0 62 ± 11 20.0 48.2 38.9 30.7 10.1 18.9 ± 2.0
Values are expressed as mean ± SD, median (inter quartile ranges), and percentage of frequencies. This analysis included a subgroup of 448 patients with normal renal function, defined as described in text and Table 1. Other abbreviations as in Table 1. a p b 0.001, vs. patients with serum creatinine levels b 1.2 mg/dL, UAE b 30 mg/day and eGFR ≥60 mL/min/1.73 m2, respectively. b p b 0.01.
the first to show that endothelial vasomotor dysfunction in systemic arteries may precede the development of renal dysfunction in patients with normal renal function.
The underlying mechanism to explain how endothelial dysfunction in systemic arteries causes renal dysfunction remains unclear. In addition to its vasodilatory action, endothelium-derived nitric oxide
Fig. 1. Odds ratio (OR) and 95% confidence intervals (95% CI) predicting progression from normal renal function to early stage renal dysfunction during 12 months follow-up in 448 patients with CAD and normal renal function at baseline, calculated by multivariate logistic regression analysis using flow-mediated dilation (FMD) in the brachial artery, age, diabetes, hypertension and C-reactive protein (CRP) levels as the independent variables. These independent variables were included because they had a significant difference between patients with and without development of the early stage renal dysfunction (Table 4). (A) progression to serum creatinine levels ≥ 1.2 mg/dL, (B) progression to urine albumin excretion (UAE) ≥30 mg/day, and (C) progression to eGFR b 60 mL/min/1.73 m2. ⁎p b 0.05. †p b 0.01.
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
also prevents inflammation, cell proliferation and fibrosis, processes that are closely related to glomerular, arteriolar and tubulointerstitial renal damage [9]. Moreover, evidence from animal experiments shows that systemic administration of a nitric oxide synthase inhibitor induces renal vasoconstriction and injury that is characterized by glomerulosclerosis, interstitial fibrosis, arterial and arteriolar lesions and cellular infiltration in the tubulointerstitial areas of the kidney [24,25]. Endothelial vasomotor dysfunction in the brachial artery may therefore reflect a decrease in nitric oxide activity in the renal vasculature that leads to renal dysfunction. These results indicate that endothelial dysfunction in systemic arteries predisposes to the development of renal dysfunction. And, these findings suggest endothelial dysfunction in systemic arteries is a common denominator of the early stages of atherosclerotic vascular dysfunction and renal dysfunction, both of which are strong risk factors for future cardiovascular events [1,2,26]. In a cross-sectional study of hypertensive patients, Perticone et al. [16] showed a significant association between the decrease in eGFR and impairment in endothelium-dependent dilator responses in microvessels of the forearm. These findings are in agreement with the results of our cross-sectional analyses and suggest a strong association exists between renal dysfunction and endothelial vasomotor dysfunction, irrespective of vessel size. Our study had some limitations. The present study arbitrarily chose cut-off levels of serum creatinine levels and UAE for criteria of normal or decreased renal function according to previous reports [1,2,6,20,21]. On the basis of eGFR [1], most of the present patients meeting the criteria of early stage renal dysfunction had an increase in CKD stage from stage 1 or 2 to stage 3 during the follow-up period. Therefore, cut-off levels of serum creatinine levels and UAE used in the present study seemed to be acceptable. It remains undetermined whether endothelial dysfunction in the brachial artery may be related to future end-stage renal disease for longer time periods. It also remains unclear whether endothelial dysfunction in the brachial artery is related to early renal dysfunction in the general population. A large prospective trial incorporating medications which improve endothelial function is therefore required in order to more precisely assess the role of endothelial dysfunction in systemic arteries in the pathogenesis of renal dysfunction progression. In conclusion, endothelial vasomotor dysfunction in the brachial artery is independently associated with renal dysfunction, and also is an independent predictor for the development of early renal dysfunction in patients with CAD. Measurement of FMD is therefore a useful marker for risk stratification of future renal dysfunction.
187
15012222” from the Ministry of Education, Culture, Sports, Science, and Technology, Health and Labor Sciences Research Grants for Comprehensive Research on Aging and Health (H15-Choju-012), Tokyo, Japan. The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [27].
Acknowledgement This study was supported by grants-in-aid for (B)(2)-15390244 and (B)-19390209, Priority Areas (C) “Medical Genome Science
Table 5 C-statistics for logistic models predicting development of early renal dysfunction during 12 month follow-up using conventional risk factors with or without FMD in patients with CAD and normal baseline renal function.
Creatinine ≥1.2 mg/dL UAE ≥30 mg/day eGFR b60 mL/min/1.73 m2
Conventional risks without FMD
Conventional risks with FMD
p-valuea
0.77 0.85 0.70
0.82 0.90 0.79
0.03 0.01 0.01
Data mean area under the ROC curve (AUC). Conventional risks included age, hypertension, diabetes mellitus, and CRP that had a significant difference between patients with and without development of early renal dysfunction, as shown in Table 4. Abbreviations as in Table 1. a p-value for the differences in AUC between conventional risks with and without FMD.
Fig. 2. Comparison of ROC curves for progression from normal renal function to early stage renal dysfunction during 12 months follow-up in 448 patients with CAD and normal renal function at baseline. (A) ROC curves for progression to serum creatinine levels ≥1.2 mg/dL. (B) ROC curves for progression to UAE ≥30 mg/day. (C) ROC curves for progression to eGFR b60 mL/min/1.73 m2. The curves are based on logistic models using covariables including age, diabetes, hypertension, and C-reactive protein (CRP) levels with or without FMD. Solid lines (blue) indicate covariates including FMD levels, and dashed lines (red) indicate the covariates without FMD levels. AUC, area under the ROC curve.
188
T. Nakamura et al. / International Journal of Cardiology 148 (2011) 183–188
References [1] Gupta R, Birnbaum Y, Uretsky BF. The renal patients with coronary artery disease. Current concepts and dilemmas. J Am Coll Cardiol 2004;44:1343–53. [2] Schiffrin EL, Lipman ML, Mann JF. Chronic kidney disease: effects on the cardiovascular system. Circulation 2007;116:85–97. [3] Jafar TH, Stark PC, Schmid CH, et al. AIPRD Study Group. Progression of chronic kidney disease: the role of blood pressure control, proteinuria, and angiotensinconverting enzyme inhibition: a patient-level meta-analysis. Ann Intern Med 2003;139:244–52. [4] Mann JF, Schmieder RE, McQueen M, et al. ONTARGET investigators. Renal outcomes with telmisartan, ramipril, or both, in people at high vascular risk (the ONTARGET study): a multicentre, randomised, double-blind, controlled trial. Lancet 2008;372:547–53. [5] Sukhija R, Bursac Z, Kakar P, et al. Effect of statins on the development of renal dysfunction. Am J Cardiol 2008;101:975–9. [6] Levey AS, Eckardt K, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from kidney disease: improving global outcomes (KDIGO). Kidney Int 2005;67:2089–100. [7] Taal MW, Brenner BM. Predicting initiation and progression of chronic kidney disease: developing renal risk scores. Kidney Int 2006;70:1694–705. [8] Deckert T, Feldt-Rasmussen B, Borch-Johnsen K, Jensen T, Kofoed-Enevoldsen A. Albuminuria reflects widespread vascular damage. The Steno hypothesis. Diabetologia 1999;32:219–26. [9] Ochodnický P, de Zeeuw D, Henning RH, Kluppel CA, van Dokkum RP. Endothelial function predicts the development of renal damage after combined nephrectomy and myocardial infarction. J Am Soc Nephrol 2006;17:S49–52. [10] King AJ, Brenner BM. Endothelium-derived vasoactive factors and the renal vasculature. Am J Physiol 1991;29:R653–62. [11] Kang DH, Kanellis J, Hugo C, et al. Role of the microvascular endothelium in progressive renal disease. J Am Soc Nephrol 2003;13:806–16. [12] Anderson TJ, Uehata A, Gerhard MD, et al. Close relation of endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol 1995;26:1235–41. [13] Stam F, van Guldener C, Becker A, et al. Endothelial dysfunction contributes to renal function-associated cardiovascular mortality in a population with mild renal insufficiency: the Hoorn study. J Am Soc Nephrol 2006;17:537–45.
[14] Ochodnicky P, Vettoretti S, Henning RH, Buikema H, Van Dokkum RP, de Zeeuw D. Endothelial dysfunction in chronic kidney disease: determinant of susceptibility to end-organ damage and therapeutic response. J Nephrol 2006;19:246–58. [15] Zoccali C. The endothelium as a target in renal diseases. J Nephrol 2007;20(suppl 12): S39–44. [16] Perticone F, Maio R, Tripepi G, Zoccali C. Endothelial dysfunction and mild renal insufficiency in essential hypertension. Circulation 2004;110:821–5. [17] van der Harst P, Smilde TD, Buikema H, et al. Vascular function and mild renal impairment in stable coronary artery disease. Arterioscler Thromb Vasc Biol 2006;26:379–84. [18] Kugiyama K, Motoyama T, Doi H, et al. Improvement of endothelial vasomotor dysfunction by treatment with alpha-tocopherol in patients with high remnant lipoproteins levels. J Am Coll Cardiol 1999;33:1512–8. [19] Nakamura T, Takano H, Umetani K, et al. Remnant lipoproteinemia is a risk factor for endothelial vasomotor dysfunction and coronary artery disease in metabolic syndrome. Atherosclerosis 2005;181:321–7. [20] Couchoud C, Pozet N, Labeeuw M, Pouteil-Noble C. Screening early renal failure: cut-off values for serum creatinine as an indicator of renal impairment. Kidney Int 1999;55:1878–84. [21] Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 1992;38:1933–53. [22] Cottone S, Lorito MC, Riccobene R, et al. Oxidative stress, inflammation and cardiovascular disease in chronic renal failure. J Nephrol 2008;21:175–9. [23] Oberg BP, McMenamin E, Lucas FL, et al. Increased prevalence of oxidant stress and inflammation in patients with moderate to severe chronic kidney disease. Kidney Int 2004;65:1009–16. [24] Jover B, Mimran A. Nitric oxide inhibition and renal alterations. J Cardiovasc Pharmacol 2001;38:S65–70. [25] Fujihara CK, Michellazzo SM, De-Nucci G, Zatz R. Sodium excess aggravates hypertension and renal parenchymal injury in rats with chronic NO inhibition. Am J Physiol 1994;266:F697–705. [26] Brunner H, Cockcroft JR, Deanfield J, et al. Endothelial function and dysfunction. Part II: association with cardiovascular risk factors and diseases. J Hypertens 2005;23:233–46. [27] Coats AJ. Ethical authorship and publishing. Int J Cardiol 2009;131:149–50.