- Email: [email protected]
Subclinical carotid atherosclerosis and triglycerides predict the incidence of chronic kidney disease in the Japanese general population: Results from the Kyushu and Okinawa Population Study (KOPS)
Atherosclerosis 238 (2015) 207e212
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Subclinical carotid atherosclerosis and triglycerides predict the incidence of chronic kidney disease in the Japanese general population: Results from the Kyushu and Okinawa Population Study (KOPS) Motohiro Shimizu a, Norihiro Furusyo a, *, Fujiko Mitsumoto a, Koji Takayama a, Kazuya Ura a, Satoshi Hiramine a, Hiroaki Ikezaki a, Takeshi Ihara a, Haru Mukae a, Eiichi Ogawa a, Kazuhiro Toyoda a, Mosaburo Kainuma a, Masayuki Murata a, Jun Hayashi a, b a b
Department of General Internal Medicine, Kyushu University Hospital, Fukuoka, Japan Kyushu General Internal Medicine Center, Haradoi Hospital, Fukuoka, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 October 2014 Received in revised form 28 November 2014 Accepted 2 December 2014 Available online 9 December 2014
Objective: To examine whether or not subclinical atherosclerosis independently predicts the incidence of chronic kidney disease (CKD) in the Japanese general population. Methods: This study is part of the Kyushu and Okinawa Population Study (KOPS), a survey of vascular events associated with lifestylerelated diseases. Participants who attended both baseline (2004e2007) and follow-up (2009e2012) examinations were eligible. The common carotid intima-media thickness (IMT) was assessed for each participant at baseline. The end point was the incidence of CKD, defined as an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2 during the follow-up of participants without CKD at baseline. Results: During the five-year follow-up, 224 of the 1824 participants (12.3%) who developed CKD had higher carotid IMT (0.74 ± 0.22 vs. 0.65 ± 0.14 mm, P < 0.001), higher triglycerides (1.6 ± 0.8 vs. 1.3 ± 0.7 mmol/L, P < 0.001), and lower high density lipoprotein cholesterol (1.5 ± 0.4 vs. 1.6 ± 0.4 mmol/ L, P < 0.001) at baseline than those who did not. In logistic regression analysis adjusted for significant covariates, eGFR (Odds ratio [OR] 0.83, 95% confidence interval (CI) 0.80e0.85, P < 0.001), carotid IMT (0.10 mm increase: OR 1.17, 95% CI 1.04e1.33, P ¼ 0.010), and triglycerides (OR 1.35, 95% CI 1.06e1.73, P ¼ 0.015) at baseline were independent predictors for the development of CKD. Conclusions: Higher carotid IMT and hypertriglyceridemia were independently associated with the development of CKD in the population studied. © 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Atherosclerosis Carotid intima-media thickness Triglycerides Chronic kidney disease (CKD) Estimated glomerular filtration rate (eGFR)
1. Introduction The incidence and prevalence of chronic kidney disease (CKD) are increasing [1e3], and it has been recognized as a medical, social, and economic problem worldwide. A mild decline of kidney function increases the risk of cardiovascular disease in the general population [4,5]. Therefore, the early detection and aggressive
* Corresponding author. Department of General Internal Medicine, Kyushu University Hospital, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. E-mail addresses: [email protected] (M. Shimizu), furusyo@ gim.med.kyushu-u.ac.jp (N. Furusyo). http://dx.doi.org/10.1016/j.atherosclerosis.2014.12.013 0021-9150/© 2014 Elsevier Ireland Ltd. All rights reserved.
modification of risk factors for the decline of kidney function are important. Aging, hypertension, and diabetes mellitus are the most common risk factors for the development of CKD [6,7]. The importance of dyslipidemia management is also emphasized in recent clinical CKD guidelines [8]. They are also the traditional risk factors of atherosclerotic cardiovascular diseases [9]; however, the relationship between atherosclerosis and kidney function has not been well elucidated. In addition, the impact of dyslipidemia on the development of incident CKD has also not yet been clarified, especially in the general population [10]. Carotid arterial intima-media thickness (IMT), a marker of subclinical atherosclerosis, is a strong predictor of future
208
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
cardiovascular events [11]. One previous study reported the correlation of carotid IMT to the incidence of CKD in a populationbased prospective study [12], but this study was limited to elderly Caucasians. Based on these findings, we hypothesized that traditional risk factors might be related to the decline of kidney function through atherosclerosis. Thus, in the present study, we examined whether or not carotid IMT independently predicts the incidence of CKD in the Japanese population. Additionally, we evaluated the impact of dyslipidemia on the development of CKD. 2. Materials and methods 2.1. Study population This study is part of the Kyushu and Okinawa Population Study (KOPS) survey of vascular events associated with lifestyle-related diseases. During the initial phase (2004e2007), 10,091 participants aged 18 were recruited from the community-dwelling population of the Kyushu and Okinawa areas of Japan. After excluding 703 participants due to a self-reported history of cardiovascular disease (n ¼ 222), refusal to undergo carotid ultrasonography (n ¼ 84), incomplete laboratory data (n ¼ 250), or high triglyceride levels (4.5 mmol/L) as required for the calculation of low density lipoprotein (LDL)-cholesterol using the Friedewald formula [13] (n ¼ 147), the data of 9388 participants was available for the baseline analysis of the present study. Follow-up assessments were performed for 2245 participants (23.9%) five years after the baseline investigation (2009e2012). The study design was approved by the Kyushu University Hospital Ethics Committee, and written informed consent was obtained from each participant prior to the examination. Some of the data from the KOPS were published previously [14e20]. The study was conducted in accordance with the principles of the Helsinki Declaration of 1975, as revised in 2000. 2.2. Anthropometric measurement and questionnaire Anthropometric measurements were performed at both baseline and follow-up with each participant wearing indoor clothing and without shoes. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared. Systolic blood pressure and diastolic blood pressure were measured two times with a oneminute interval on the right arm in the sitting position with an automated sphygmomanometer (HEM-780, Omron Healthcare, Kyoto, Japan [21]) after a five-minute rest. The mean of the two measurements was used for analysis. Each participant completed a self-administrated questionnaire that included smoking status, alcohol consumption, medical history, and use of drugs. The questionnaire was checked for unfilled or inconsistent answers, first by nurses and again by our staff physicians. 2.3. Laboratory measurements Blood samples were collected at baseline and follow-up after an overnight fast of at least 8 h. Plasma glucose concentration was measured using the hexokinase-glucose-6-phosphate dehydrogenase method, and the HbA1c level was measured by immunoassay of fresh whole blood samples. The HbA1c levels are expressed as US National Glycohemoglobin Standardization Program format levels (%). At baseline, the serum levels of creatinine, total cholesterol, triglycerides, and high density lipoprotein (HDL) cholesterol were determined enzymatically, and the LDL-cholesterol level was calculated using the Friedewald formula. In contrast, at follow-up the LDL-cholesterol level was determined by a direct method.
Details of these laboratory measurements were described previously [15e17,19].
2.4. Ultrasonographic measurement The baseline carotid IMT was measured by ultrasound, as previously described [14,17]. Briefly, the participants were supine with a slight hyperextension and rotation of the neck in the direction opposite the probe. Carotid artery lesions were measured using high resolution B-mode ultrasonography with a 7.5 MHz linear array probe (UF-4300R®, Fukuda Denshi Co., Ltd, Tokyo, Japan) by the well-trained physicians of our department. The intra-class correlation coefficients within and between observers were 0.90 and 0.91, respectively. The mean-IMT was estimated at 20, 25, and 30 mm proximal to the bifurcation of flow at the far wall of both the right and left sides and was calculated as the mean value for the six points.
2.5. Definition of end point The primary end point was the incidence of CKD, defined as an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2 at follow-up for participants without CKD at baseline. The eGFR was calculated using the Modification of Diet in Renal Disease study equation modified for Japanese subjects: eGFR (mL/min/ 1.73 m2) ¼ 194 * age0.287 * serum creatinine (mg/dL)1.094 (if female * 0.739) [22]. The secondary end point was the change in eGFR during the five years between baseline and follow-up. Participants who died or who moved away from the study areas were excluded from the prospective analysis.
Table 1 Baseline clinical characteristics. Baseline eGFR 60 mL/min/ 1.73 m2
Age (years) Male (%) Body mass index (kg/m2) Smoking habit (%) Habitual alcohol intake (%) Medication for hypertension (%) Medication for diabetes (%) Medication for dislipidemia (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Carotid IMT (mm) eGFR (mL/min/1.73 m2)
P value <60 mL/min/ 1.73 m2
(N ¼ 7885)
(N ¼ 1503)
57.5 ± 13.8 34.4 23.5 ± 3.5 15.1 38.6 18.2
70.0 ± 10.0 36.1 24.6 ± 3.5 8.1 31.2 35.7
<0.001 0.213 <0.001 <0.001 <0.001 <0.001
2.9 6.7
4.2 14.5
0.028 <0.001
127.1 ± 20.2
132.1 ± 17.1
<0.001
75.5 ± 12.5
75.8 ± 11.2
5.5 ± 1.4
5.9 ± 1.6
<0.001
5.5 ± 0.7 1.3 ± 0.7 1.6 ± 0.4 3.1 ± 0.8 0.68 ± 0.16 80.0 ± 14.0
5.6 ± 0.7 1.5 ± 0.7 1.5 ± 0.4 3.1 ± 0.8 0.76 ± 0.17 51.8 ± 7.9
<0.001 <0.001 <0.001 0.041 <0.001 <0.001
0.347
Data shown as mean ± standard deviation or percentage. Overall p values were calculated by unpaired t-test or chi-square test. HDL, high density lipoprotein; LDL, low density lipoprotein; IMT, intima-media thickness. eGFR, estimated glomerular filtration rate.
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
2.6. Statistical analysis All data are expressed as means ± standard deviation (SD) or as a percentage. For comparisons of participants with/without CKD at baseline or who did/did not develop CKD during the follow-up, unpaired Student's t-test was used to compare mean values and the c2-test was used to evaluate differences in prevalence rates. The odds ratio (OR) and the 95% confidence interval (CI) were calculated using multiple logistic regression analysis after adjustment for known covariates. Analysis of covariance was performed to detect differences among subgroups after adjustment for confounding factors, and the Bonferroni test was used for multiple pairwise comparisons. The statistical calculations were performed using the computer software package SPSS version 22.0 (SPSS Inc., IBM, Somers, NY). A two-tailed P value <0.05 was considered to be statistically significant. 3. Results
209
level, and carotid IMT level (Supplemental Table 1). During the five years of follow-up, 224 of the 1824 participants (12.3%) developed CKD. They had higher carotid IMT, higher triglycerides, and lower HDL-cholesterol at baseline than those who did not (Table 3). Age, BMI, smoking status, the use of medications for hypertension and dyslipidemia, systolic blood pressure, fasting plasma glucose, and HbA1c at baseline were also significantly different between the two groups. Logistic regression analysis adjusted for the significant covariates found in the univariate analyses extracted baseline eGFR, baseline carotid IMT, and baseline triglycerides as independent predictors for the development of CKD (Table 4). When the known baseline covariates sex, drinking status, medication for diabetes, LDL-cholesterol, and diastolic blood pressure were forced into the model, similar results were found. In addition, we evaluated the mean change in eGFR during the five years between baseline and follow-up for subgroups classified by tertile of baseline carotid IMT (Fig. 1). The participants in the top tertile had a greater mean change in eGFR than the other groups.
3.1. Baseline characteristics The baseline age of the 3254 men (34.7%) and 6134 women (65.3%) studied ranged from 18 to 100 years (mean ± SD: 59.5 ± 14.0 years). The percentage of participants who used medications for hypertension, diabetes, or dyslipidemia was 20.4%, 3.1%, and 7.7%, respectively. The clinical characteristics of the participants with/without CKD at baseline are presented in Table 1. Significant between group differences were found in age, BMI, smoking status, drinking status, the use of medications for hypertension, diabetes, and dyslipidemia, not but for sex. Carotid IMT was also higher for participants with than without CKD; however, it was not independently related to the presence of CKD at baseline in a multivariate analysis adjusted for the significant variables identified in univariate analysis (Table 2). 3.2. Follow-up assessments Of the 7885 participants without CKD at baseline, 1824 (23.1%) had 5-year follow-up assessments. In comparison of participants with and without follow-up assessment, no significant differences were found in baseline age, sex, habitual drinking, antidiabetic or antilipidemic medication use, fasting plasma glucose level, HbA1c level, or HDL-cholesterol level; however, significant differences were found in BMI, current smoking, antihypertensive medication use, blood pressure, triglyceride levels, LDL-cholesterol level, eGFR
Table 2 Logistic regression analysis for the presence of CKD at baseline.
Age (years) Body mass index (kg/m2) Smoking habit (vs. absence) Habitual alcohol intake (vs. absence) Medication for hypertension (vs. absence) Medication for diabetes (vs. absence) Medication for dislipidemia (vs. absence) Systolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Carotid IMT (per 0.10 mm)
OR
95% CI
P value
1.08 1.08 0.85 0.86 1.26 0.95 1.42 0.99 1.13 0.66 1.46 0.82 1.06 0.96
1.07e1.09 1.05e1.10 0.66e1.10 0.73e1.02 1.06e1.50 0.61e1.48 1.14e1.78 0.989e0.997 1.06e1.21 0.57e0.77 1.32e1.61 0.66e1.03 0.96e1.16 0.91e1.01
<0.001 <0.001 0.209 0.082 0.009 0.814 0.002 0.001 <0.001 <0.001 <0.001 0.086 0.257 0.127
OR, 95% CI, and P value were calculated by logistic regression analysis. CKD, chronic kidney disease; OR, odds ratio; CI, confidence interval; HDL, high density lipoprotein; LDL, low density lipoprotein; IMT, intima-media thickness.
Table 3 Comparison of participants who did/did not develop CKD during follow-up. Follow-up eGFR
Baseline Age (years) Male (%) Body mass index (kg/m2) Smoking habit (%) Habitual alcohol intake (%) Medication for hypertension (%) Medication for diabetes (%) Medication for dislipidemia (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) eGFR (mL/min/1.73 m2) Carotid IMT (mm) Follow-up Body mass index (kg/m2) Medication for hypertension (%) Medication for diabetes (%) Medication for dislipidemia (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) eGFR (mL/min/1.73 m2)
P value
60 mL/min/ 1.73 m2
<60 mL/min/ 1.73 m2
(N ¼ 1600)
(N ¼ 224)
56.1 ± 10.9 35.1 23.6 ± 3.3 14.1 41.0 14.5
64.5 ± 9.7 34.4 24.8 ± 3.2 7.7 37.4 23.8
2.9 5.9
2.2 10.3
123.8 ± 17.1
130.1 ± 18.3
73.9 ± 11.4
74.8 ± 10.5
0.315
5.4 ± 1.3
5.8 ± 1.9
0.008
5.5 ± 0.6 1.3 ± 0.7 1.6 ± 0.4 3.1 ± 0.8 78.1 ± 11.7 0.65 ± 0.14
5.7 ± 0.8 1.6 ± 0.8 1.5 ± 0.4 3.1 ± 0.9 66.1 ± 5.8 0.74 ± 0.22
0.004 <0.001 <0.001 0.485 <0.001 <0.001
23.3 ± 3.4 23.2
24.4 ± 3.1 39.9
<0.001 <0.001
5.0 12.1
7.6 17.5
124.6 ± 16.6
131.4 ± 17.5
74.4 ± 10.6
74.9 ± 9.2
5.5 ± 1.1
5.9 ± 1.6
<0.001
5.6 ± 0.6 1.3 ± 0.9 1.7 ± 0.4 3.2 ± 0.8 76.3 ± 11.4
5.8 ± 0.7 1.6 ± 1.3 1.5 ± 0.4 3.3 ± 0.8 54.8 ± 5.4
0.003 <0.001 <0.001 0.004 <0.001
<0.001 0.840 <0.001 0.016 0.383 0.001 0.589 0.021 <0.001
0.102 0.025 <0.001 0.476
Data shown as mean ± standard deviation or percentage. P values calculated by unpaired t-test or chi-square test between participants who did/did not develope CKD in the follow-up phase. CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HDL, high density lipoprotein; LDL, low density lipoprotein; IMT, intima-media thickness.
210
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
Table 4 Logistic regression analysis for the development of CKD during follow-up.
Model 1a Baseline Baseline Baseline Model 2b Baseline Baseline Baseline
OR
95% CI
P value
eGFR (mL/min/1.73 m2) Carotid IMT (per 0.10 mm) Triglycerides (mmol/L)
0.83 1.17 1.35
0.80e0.85 1.04e1.33 1.06e1.73
<0.001 0.010 0.015
eGFR (mL/min/1.73 m2) Carotid IMT (per 0.10 mm) Triglycerides (mmol/L)
0.84 1.19 1.41
0.81e0.87 1.05e1.36 1.09e1.83
<0.001 0.008 0.008
OR, 95% CI, and P value were calculated by Logistic regression analysis. CKD, chronic kidney disease; OR, odds ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; IMT, intima-media thickness. a Adjusted for baseline variables age, body mass index, smoking status, medication for hypertension, medication for dyslipidemia, systolic blood pressure, glucose, HbA1c, and HDL-cholesterol. b Adjusted for baseline variables sex, drinking status, medication for diabetes, LDL-cholesterol, and diastolic blood pressure (in addition to the adjustments made for model 1).
Fig. 1. The decrease in eGFR by the tertile of baseline carotid IMT during the five years between baseline and follow-up. Data are shown as adjusted mean values of the change in eGFR and 95% confidence intervals in subgroups classified by tertile of baseline carotid IMT. The change in eGFR was adjusted for the baseline variables age, body mass index, smoking status, medication for hypertension, medication for dyslipidemia, systolic blood pressure, fasting plasma glucose, HbA1c, triglycerides, HDLcholesterol, and eGFR. Intergroup differences were calculated by the Bonferroni test. IMT, intima-media thickness; eGFR, estimated glomerular filtration rate; HDL, high density lipoprotein.
4. Discussion The main finding of the present 5-year prospective cohort study of a Japanese population is that carotid IMT is a significant predictor for the development of CKD and the decline in eGFR, independently of the traditional risk factors. Moreover, triglycerides were also significantly associated with the development of CKD. These results suggest that the evaluation of carotid atherosclerosis and serum triglyceride levels may be useful for the early detection of CKD in relatively healthy individuals without a history of cardiovascular disease. In this analysis, carotid IMT was an independent predictor for the development of CKD; while the traditional risk factors of age, BMI, systolic blood pressure, fasting plasma glucose, and HbA1c were not related in multivariate analysis that included carotid IMT. Chonchol et al. previously reported that carotid IMT, systolic blood
pressure, HbA1c, and baseline eGFR were significant determinants of the development of CKD [12]. We believe that the different characteristics of the population we studied, in which healthy participants were the majority, may be responsible for these discrepancies. The population of Chonchol's study was aged >55 years and included patients with a history of cardiovascular disease (about 10%). The mean levels of systolic blood pressure (139 mmHg), HbA1c (5.8%), and carotid IMT (0.78 mm) were also higher in their study. Therefore, we feel that the impact of blood pressure and HbA1c may be relatively low and that carotid IMT would be a superior indicator of the future development of CKD for people with low cardiovascular risk. The presence of carotid atherosclerosis could indicate nephrosclerosis. A positive relationship between carotid IMT and nephrosclerosis was reported among kidney transplant donors [23]. Thus, for relatively healthy people, traditional risk factors may induce the decline of kidney function through the progression of nephrosclerosis. This is a possible explanation as to why carotid atherosclerosis is useful for predicting future CKD and the decline in eGFR, over and above the traditional risk factors. Carotid IMT was not significantly correlated with the presence of CKD in the baseline multivariate analysis of this study. Some previous cross-sectional studies have shown a significant relationship between carotid IMT and eGFR in patients with hypertension [24,25], with impaired glucose metabolism [24,26], or with coronary artery disease [27]. In contrast, among relatively healthy subjects, this relationship was not found in univariate [28,29] or multivariate analyses [30e32], which is similar to our results. The significant cross-sectional correlation between carotid IMT and eGFR may be limited to subjects with advanced atherosclerosis and/or kidney dysfunction. In the present study, the baseline triglyceride levels were positively associated with the development of CKD, independently of carotid IMT. A few community-based studies have shown a correlation between hypertriglyceridemia and kidney dysfunction [33,34]. Furthermore, among patients with type 2 diabetes, a relationship between hypertriglyceridemia and diabetic kidney disease with eGFR < 60 mL/min/1.73 m2 and/or microalbuminuria was shown by an international case-control study [35]. Hypertriglyceridemia is related to insulin resistance [36], which has been reported to be a risk factor for the decline of kidney function [37]. Renal accumulation of triglycerides and free fatty acids has also been suggested to be associated with kidney injury and dysfunction [38]. This, taken together with our findings, indicates that hypertriglyceridemia contributes to the development of CKD through insulin resistance (and compensatory hyperinsulinemia) or lipotoxity, independently of atherosclerosis. Several limitations of this study bear mention. First, the possibility of selection bias because of the exclusion of the participants who were not available for follow-up assessment needs to be considered when generalizing the present findings. Follow-up assessments were done for only 23.1%. Second, the diagnosis of CKD was based on a single examination, which could lead to misclassification. Third, data regarding microalbuminuria and proteinuria were not collected, and no information about the etiology of the CKD was obtained. Fourth, non-traditional risk factors associated with atherosclerosis or kidney dysfunction, such as inflammatory markers, insulin and insulin resistance markers, and homocysteine, were not explored in the present study. Finally, we could not evaluate the relationship between the 5-year change in carotid IMT and that of eGFR. The impact of both time-related and treatmentinduced changes of carotid atherosclerosis on the decline of kidney function should be evaluated in future research. Despite the above limitations, we believe that our findings from this large community-based sample will contribute the prevention of CKD in
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
healthy individuals. 5. Conclusions We found that subclinical carotid atherosclerosis is predictive of the future incidence of CKD in a Japanese general population. Moreover, serum triglyceride levels were also independently associated with the development of CKD. Further research should focus on which groups in the general population would benefit most from the evaluation of carotid atherosclerosis. In addition, more research is needed to investigate whether or not aggressive treatment for cardiovascular risk factors among persons with subclinical atherosclerosis can improve renal prognosis thorough the prevention or reduction of atherosclerosis. Financial support This study was supported by the Japan Multi-institutional Collaborative Cohort Study (J-MICC Study), a Scientific Support Program for Cancer Research Grant-in-Aid for Scientific Research on Innovative Areas (No. 221S001), and a Grant-in-Aid for Comprehensive Research of the 21st Century COE Program from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Conflict of interest This paper has not been published or presented elsewhere in part or in its entirety and is not under consideration by another journal. M.S, F.M, K.T, K.U, S.H, H.I, T.I, H.M, E.O, K.T, M.K, M.M, and J.H. have nothing to declare. N.F. has received support from Daiichi Sankyo Healthcare Co., Ltd., Tokyo, Japan, Chugai Pharmaceutical Co., Ltd., Tokyo, Japan, MSD Ltd, Tokyo, Japan and Mitsubishi Tanabe Pharma, Osaka, Japan. The remaining authors disclose no conflicts. Acknowledgments We are grateful to Drs. Kyoko Okada, Hiroaki Taniai, Hachiro Ohnishi, Tsunehisa Koga, Kunimitsu Eiraku, Takeo Hayashi, Yuji Harada, Sakiko Hayasaki, Ayaka Komori, Rinne Sakemi, Sho Yamasaki, Azusa Hatashima, Yoshifumi Kato, Yuki Tanaka, and Masaru Sakiyama from our department for their assistance. We also thank Ms. Setsuko Nagata and Takako Nagamine for their invaluable assistance. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2014.12.013. References [1] S.I. Hallan, J. Coresh, B.C. Astor, et al., International comparison of the relationship of chronic kidney disease prevalence and ESRD risk, J. Am. Soc. Nephrol. 17 (2006) 2275e2284. [2] J. Coresh, E. Selvin, L.A. Stevens, et al., Prevalence of chronic kidney disease in the United States, J. Am. Med. Assoc. 298 (2007) 2038e2047. [3] E. Imai, M. Horio, T. Watanabe, et al., Prevalence of chronic kidney disease in the Japanese general population, Clin. Exp. Nephrol. 13 (2009) 621e630. [4] P. Muntner, J. He, L. Hamm, et al., Renal insufficiency and subsequent death resulting from cardiovascular disease in the United States, J. Am. Soc. Nephrol. 13 (2002) 745e753. [5] C. Meisinger, A. Doring, H. Lowel, Chronic kidney disease and risk of incident myocardial infarction and all-cause and cardiovascular disease mortality in middle-aged men and women from the general population, Eur. Heart J. 27 (2006) 1245e1250. [6] N.B. Shulman, C.E. Ford, W.D. Hall, et al., Prognostic value of serum creatinine and effect of treatment of hypertension on renal function. Results from the hypertension detection and follow-up program. The Hypertension Detection
211
and Follow-up Program Cooperative Group, Hypertension 13 (1989) I80eI93. [7] K. Yamagata, K. Ishida, T. Sairenchi, et al., Risk factors for chronic kidney disease in a community-based population: a 10-year follow-up study, Kidney Int. 71 (2007) 159e166. [8] M. Tonelli, C. Wanner, Lipid management in chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2013 clinical practice guideline, Ann. Intern. Med. 160 (2014) 182e189. [9] M. Shimizu, J. Ishikawa, K. Eguchi, et al., Association of an abnormal blood glucose level and morning blood pressure surge in elderly subjects with hypertension, Am. J. Hypertens. 22 (2009) 611e616. [10] A. Cases, E. Coll, Dyslipidemia and the progression of renal disease in chronic renal failure patients, Kidney Int. Suppl. (2005) S87eS93. [11] M.W. Lorenz, H.S. Markus, M.L. Bots, et al., Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and metaanalysis, Circulation 115 (2007) 459e467. [12] M. Chonchol, H. Gnahn, D. Sander, Impact of subclinical carotid atherosclerosis on incident chronic kidney disease in the elderly, Nephrol. Dial. Transpl. 23 (2008) 2593e2598. [13] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499e502. [14] H. Ohnishi, Y. Sawayama, N. Furusyo, et al., Risk factors for and the prevalence of peripheral arterial disease and its relationship to carotid atherosclerosis: the Kyushu and Okinawa Population Study (KOPS), J. Atheroscler. Thromb. 17 (2010) 751e758. [15] N. Furusyo, T. Koga, M. Ai, et al., Utility of glycated albumin for the diagnosis of diabetes mellitus in a Japanese population study: results from the Kyushu and Okinawa Population Study (KOPS), Diabetologia 54 (2011) 3028e3036. [16] M. Unno, N. Furusyo, H. Mukae, et al., The utility of visceral fat level by bioelectrical impedance analysis in the screening of metabolic syndrome - the results of the Kyushu and Okinawa Population Study (KOPS), J. Atheroscler. Thromb. 19 (2012) 462e470. [17] N. Furusyo, T. Koga, M. Ai, et al., Plasma glycated albumin level and atherosclerosis: results from the Kyushu and Okinawa Population Study (KOPS), Int. J. Cardiol. 167 (2013) 2066e2072. [18] T. Ihara, N. Furusyo, T. Hayashi, et al., A population-based epidemiological survey of human parvovirus B19 infection: a project of the Kyushu and Okinawa Population Study (KOPS), Arch. Virol. 158 (2013) 2465e2472. [19] H. Ikezaki, N. Furusyo, K. Okada, et al., The utility of urinary myo-inositol as a marker of glucose intolerance, Diabetes Res. Clin. Pract. 103 (2014) 88e96. [20] N. Furusyo, M. Ai, M. Okazaki, et al., Serum cholesterol and triglyceride reference ranges of twenty lipoprotein subclasses for healthy Japanese men and women, Atherosclerosis 231 (2013) 238e245. [21] A. Coleman, S. Steel, P. Freeman, et al., Validation of the Omron M7 (HEM-780E) oscillometric blood pressure monitoring device according to the British Hypertension Society protocol, Blood Press Monit. 13 (2008) 49e54. [22] S. Matsuo, E. Imai, M. Horio, et al., Revised equations for estimated GFR from serum creatinine in Japan, Am. J. Kidney Dis. 53 (2009) 982e992. [23] S. Erten, O. Gungor, S. Sen, et al., Nephrosclerosis and carotid atherosclerosis: lessons from kidney donor histology, Nephrol. (Carlton) 16 (2011) 720e724. [24] N. Ishizaka, Y. Ishizaka, E. Toda, et al., Association between chronic kidney disease and carotid intima-media thickening in individuals with hypertension and impaired glucose metabolism, Hypertens. Res. 30 (2007) 1035e1041. [25] T. Ohara, Y. Kokubo, K. Toyoda, et al., Impact of chronic kidney disease on carotid atherosclerosis according to blood pressure category: the Suita study, Stroke 44 (2013) 3537e3539. [26] H. Ito, Y. Komatsu, M. Mifune, et al., The estimated GFR, but not the stage of diabetic nephropathy graded by the urinary albumin excretion, is associated with the carotid intima-media thickness in patients with type 2 diabetes mellitus: a cross-sectional study, Cardiovasc Diabetol. 9 (2010) 18. [27] A. Lisowska, W.J. Musial, P. Lisowski, et al., Intima-media thickness is a useful marker of the extent of coronary artery disease in patients with impaired renal function, Atherosclerosis 202 (2009) 470e475. [28] P. Dasmahapatra, S.R. Srinivasan, J. Mokha, et al., Subclinical atherosclerotic changes related to chronic kidney disease in asymptomatic black and white young adults: the Bogalusa heart study, Ann. Epidemiol. 21 (2011) 311e317. [29] X. Kong, X. Jia, Y. Wei, et al., Association between microalbuminuria and subclinical atherosclerosis evaluated by carotid artery intima-media in elderly patients with normal renal function, BMC Nephrol. 13 (2012) 37. [30] L. Zhang, F. Zhao, Y. Yang, et al., Association between carotid artery intimamedia thickness and early-stage CKD in a Chinese population, Am. J. Kidney Dis. 49 (2007) 786e792. [31] S.W. Choi, H.Y. Kim, Y.H. Lee, et al., eGFR is associated with subclinical atherosclerosis independent of albuminuria: the Dong-gu Study, Atherosclerosis 212 (2010) 661e667. [32] L. Han, X. Bai, H. Lin, et al., Lack of independent relationship between agerelated kidney function decline and carotid intima-media thickness in a healthy Chinese population, Nephrol. Dial. Transpl. 25 (2010) 1859e1865. [33] P. Muntner, J. Coresh, J.C. Smith, et al., Plasma lipids and risk of developing renal dysfunction: the atherosclerosis risk in communities study, Kidney Int. 58 (2000) 293e301. [34] F. Wang, P. Ye, L. Luo, et al., Association of risk factors for cardiovascular disease and glomerular filtration rate: a community-based study of 4,925 adults in Beijing, Nephrol. Dial. Transpl. 25 (2010) 3924e3931. [35] F.M. Sacks, M.P. Hermans, P. Fioretto, et al., Association between plasma
212
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
triglycerides and high-density lipoprotein cholesterol and microvascular kidney disease and retinopathy in type 2 diabetes mellitus: a global casecontrol study in 13 countries, Circulation 129 (2014) 999e1008. [36] T. McLaughlin, F. Abbasi, K. Cheal, et al., Use of metabolic markers to identify overweight individuals who are insulin resistant, Ann. Intern Med. 139 (2003) 802e809.
[37] G. Dogra, A. Irish, D. Chan, et al., Insulin resistance, inflammation, and blood pressure determine vascular dysfunction in CKD, Am. J. Kidney Dis. 48 (2006) 926e934. [38] I.A. Bobulescu, Renal lipid metabolism and lipotoxicity, Curr. Opin. Nephrol. Hypertens. 19 (2010) 393e402.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Subclinical carotid atherosclerosis and triglycerides predict the incidence of chronic kidney disease in the Japanese general population: Results from the Kyushu and Okinawa Population Study (KOPS) Motohiro Shimizu a, Norihiro Furusyo a, *, Fujiko Mitsumoto a, Koji Takayama a, Kazuya Ura a, Satoshi Hiramine a, Hiroaki Ikezaki a, Takeshi Ihara a, Haru Mukae a, Eiichi Ogawa a, Kazuhiro Toyoda a, Mosaburo Kainuma a, Masayuki Murata a, Jun Hayashi a, b a b
Department of General Internal Medicine, Kyushu University Hospital, Fukuoka, Japan Kyushu General Internal Medicine Center, Haradoi Hospital, Fukuoka, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 October 2014 Received in revised form 28 November 2014 Accepted 2 December 2014 Available online 9 December 2014
Objective: To examine whether or not subclinical atherosclerosis independently predicts the incidence of chronic kidney disease (CKD) in the Japanese general population. Methods: This study is part of the Kyushu and Okinawa Population Study (KOPS), a survey of vascular events associated with lifestylerelated diseases. Participants who attended both baseline (2004e2007) and follow-up (2009e2012) examinations were eligible. The common carotid intima-media thickness (IMT) was assessed for each participant at baseline. The end point was the incidence of CKD, defined as an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2 during the follow-up of participants without CKD at baseline. Results: During the five-year follow-up, 224 of the 1824 participants (12.3%) who developed CKD had higher carotid IMT (0.74 ± 0.22 vs. 0.65 ± 0.14 mm, P < 0.001), higher triglycerides (1.6 ± 0.8 vs. 1.3 ± 0.7 mmol/L, P < 0.001), and lower high density lipoprotein cholesterol (1.5 ± 0.4 vs. 1.6 ± 0.4 mmol/ L, P < 0.001) at baseline than those who did not. In logistic regression analysis adjusted for significant covariates, eGFR (Odds ratio [OR] 0.83, 95% confidence interval (CI) 0.80e0.85, P < 0.001), carotid IMT (0.10 mm increase: OR 1.17, 95% CI 1.04e1.33, P ¼ 0.010), and triglycerides (OR 1.35, 95% CI 1.06e1.73, P ¼ 0.015) at baseline were independent predictors for the development of CKD. Conclusions: Higher carotid IMT and hypertriglyceridemia were independently associated with the development of CKD in the population studied. © 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Atherosclerosis Carotid intima-media thickness Triglycerides Chronic kidney disease (CKD) Estimated glomerular filtration rate (eGFR)
1. Introduction The incidence and prevalence of chronic kidney disease (CKD) are increasing [1e3], and it has been recognized as a medical, social, and economic problem worldwide. A mild decline of kidney function increases the risk of cardiovascular disease in the general population [4,5]. Therefore, the early detection and aggressive
* Corresponding author. Department of General Internal Medicine, Kyushu University Hospital, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. E-mail addresses: [email protected] (M. Shimizu), furusyo@ gim.med.kyushu-u.ac.jp (N. Furusyo). http://dx.doi.org/10.1016/j.atherosclerosis.2014.12.013 0021-9150/© 2014 Elsevier Ireland Ltd. All rights reserved.
modification of risk factors for the decline of kidney function are important. Aging, hypertension, and diabetes mellitus are the most common risk factors for the development of CKD [6,7]. The importance of dyslipidemia management is also emphasized in recent clinical CKD guidelines [8]. They are also the traditional risk factors of atherosclerotic cardiovascular diseases [9]; however, the relationship between atherosclerosis and kidney function has not been well elucidated. In addition, the impact of dyslipidemia on the development of incident CKD has also not yet been clarified, especially in the general population [10]. Carotid arterial intima-media thickness (IMT), a marker of subclinical atherosclerosis, is a strong predictor of future
208
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
cardiovascular events [11]. One previous study reported the correlation of carotid IMT to the incidence of CKD in a populationbased prospective study [12], but this study was limited to elderly Caucasians. Based on these findings, we hypothesized that traditional risk factors might be related to the decline of kidney function through atherosclerosis. Thus, in the present study, we examined whether or not carotid IMT independently predicts the incidence of CKD in the Japanese population. Additionally, we evaluated the impact of dyslipidemia on the development of CKD. 2. Materials and methods 2.1. Study population This study is part of the Kyushu and Okinawa Population Study (KOPS) survey of vascular events associated with lifestyle-related diseases. During the initial phase (2004e2007), 10,091 participants aged 18 were recruited from the community-dwelling population of the Kyushu and Okinawa areas of Japan. After excluding 703 participants due to a self-reported history of cardiovascular disease (n ¼ 222), refusal to undergo carotid ultrasonography (n ¼ 84), incomplete laboratory data (n ¼ 250), or high triglyceride levels (4.5 mmol/L) as required for the calculation of low density lipoprotein (LDL)-cholesterol using the Friedewald formula [13] (n ¼ 147), the data of 9388 participants was available for the baseline analysis of the present study. Follow-up assessments were performed for 2245 participants (23.9%) five years after the baseline investigation (2009e2012). The study design was approved by the Kyushu University Hospital Ethics Committee, and written informed consent was obtained from each participant prior to the examination. Some of the data from the KOPS were published previously [14e20]. The study was conducted in accordance with the principles of the Helsinki Declaration of 1975, as revised in 2000. 2.2. Anthropometric measurement and questionnaire Anthropometric measurements were performed at both baseline and follow-up with each participant wearing indoor clothing and without shoes. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared. Systolic blood pressure and diastolic blood pressure were measured two times with a oneminute interval on the right arm in the sitting position with an automated sphygmomanometer (HEM-780, Omron Healthcare, Kyoto, Japan [21]) after a five-minute rest. The mean of the two measurements was used for analysis. Each participant completed a self-administrated questionnaire that included smoking status, alcohol consumption, medical history, and use of drugs. The questionnaire was checked for unfilled or inconsistent answers, first by nurses and again by our staff physicians. 2.3. Laboratory measurements Blood samples were collected at baseline and follow-up after an overnight fast of at least 8 h. Plasma glucose concentration was measured using the hexokinase-glucose-6-phosphate dehydrogenase method, and the HbA1c level was measured by immunoassay of fresh whole blood samples. The HbA1c levels are expressed as US National Glycohemoglobin Standardization Program format levels (%). At baseline, the serum levels of creatinine, total cholesterol, triglycerides, and high density lipoprotein (HDL) cholesterol were determined enzymatically, and the LDL-cholesterol level was calculated using the Friedewald formula. In contrast, at follow-up the LDL-cholesterol level was determined by a direct method.
Details of these laboratory measurements were described previously [15e17,19].
2.4. Ultrasonographic measurement The baseline carotid IMT was measured by ultrasound, as previously described [14,17]. Briefly, the participants were supine with a slight hyperextension and rotation of the neck in the direction opposite the probe. Carotid artery lesions were measured using high resolution B-mode ultrasonography with a 7.5 MHz linear array probe (UF-4300R®, Fukuda Denshi Co., Ltd, Tokyo, Japan) by the well-trained physicians of our department. The intra-class correlation coefficients within and between observers were 0.90 and 0.91, respectively. The mean-IMT was estimated at 20, 25, and 30 mm proximal to the bifurcation of flow at the far wall of both the right and left sides and was calculated as the mean value for the six points.
2.5. Definition of end point The primary end point was the incidence of CKD, defined as an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2 at follow-up for participants without CKD at baseline. The eGFR was calculated using the Modification of Diet in Renal Disease study equation modified for Japanese subjects: eGFR (mL/min/ 1.73 m2) ¼ 194 * age0.287 * serum creatinine (mg/dL)1.094 (if female * 0.739) [22]. The secondary end point was the change in eGFR during the five years between baseline and follow-up. Participants who died or who moved away from the study areas were excluded from the prospective analysis.
Table 1 Baseline clinical characteristics. Baseline eGFR 60 mL/min/ 1.73 m2
Age (years) Male (%) Body mass index (kg/m2) Smoking habit (%) Habitual alcohol intake (%) Medication for hypertension (%) Medication for diabetes (%) Medication for dislipidemia (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Carotid IMT (mm) eGFR (mL/min/1.73 m2)
P value <60 mL/min/ 1.73 m2
(N ¼ 7885)
(N ¼ 1503)
57.5 ± 13.8 34.4 23.5 ± 3.5 15.1 38.6 18.2
70.0 ± 10.0 36.1 24.6 ± 3.5 8.1 31.2 35.7
<0.001 0.213 <0.001 <0.001 <0.001 <0.001
2.9 6.7
4.2 14.5
0.028 <0.001
127.1 ± 20.2
132.1 ± 17.1
<0.001
75.5 ± 12.5
75.8 ± 11.2
5.5 ± 1.4
5.9 ± 1.6
<0.001
5.5 ± 0.7 1.3 ± 0.7 1.6 ± 0.4 3.1 ± 0.8 0.68 ± 0.16 80.0 ± 14.0
5.6 ± 0.7 1.5 ± 0.7 1.5 ± 0.4 3.1 ± 0.8 0.76 ± 0.17 51.8 ± 7.9
<0.001 <0.001 <0.001 0.041 <0.001 <0.001
0.347
Data shown as mean ± standard deviation or percentage. Overall p values were calculated by unpaired t-test or chi-square test. HDL, high density lipoprotein; LDL, low density lipoprotein; IMT, intima-media thickness. eGFR, estimated glomerular filtration rate.
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
2.6. Statistical analysis All data are expressed as means ± standard deviation (SD) or as a percentage. For comparisons of participants with/without CKD at baseline or who did/did not develop CKD during the follow-up, unpaired Student's t-test was used to compare mean values and the c2-test was used to evaluate differences in prevalence rates. The odds ratio (OR) and the 95% confidence interval (CI) were calculated using multiple logistic regression analysis after adjustment for known covariates. Analysis of covariance was performed to detect differences among subgroups after adjustment for confounding factors, and the Bonferroni test was used for multiple pairwise comparisons. The statistical calculations were performed using the computer software package SPSS version 22.0 (SPSS Inc., IBM, Somers, NY). A two-tailed P value <0.05 was considered to be statistically significant. 3. Results
209
level, and carotid IMT level (Supplemental Table 1). During the five years of follow-up, 224 of the 1824 participants (12.3%) developed CKD. They had higher carotid IMT, higher triglycerides, and lower HDL-cholesterol at baseline than those who did not (Table 3). Age, BMI, smoking status, the use of medications for hypertension and dyslipidemia, systolic blood pressure, fasting plasma glucose, and HbA1c at baseline were also significantly different between the two groups. Logistic regression analysis adjusted for the significant covariates found in the univariate analyses extracted baseline eGFR, baseline carotid IMT, and baseline triglycerides as independent predictors for the development of CKD (Table 4). When the known baseline covariates sex, drinking status, medication for diabetes, LDL-cholesterol, and diastolic blood pressure were forced into the model, similar results were found. In addition, we evaluated the mean change in eGFR during the five years between baseline and follow-up for subgroups classified by tertile of baseline carotid IMT (Fig. 1). The participants in the top tertile had a greater mean change in eGFR than the other groups.
3.1. Baseline characteristics The baseline age of the 3254 men (34.7%) and 6134 women (65.3%) studied ranged from 18 to 100 years (mean ± SD: 59.5 ± 14.0 years). The percentage of participants who used medications for hypertension, diabetes, or dyslipidemia was 20.4%, 3.1%, and 7.7%, respectively. The clinical characteristics of the participants with/without CKD at baseline are presented in Table 1. Significant between group differences were found in age, BMI, smoking status, drinking status, the use of medications for hypertension, diabetes, and dyslipidemia, not but for sex. Carotid IMT was also higher for participants with than without CKD; however, it was not independently related to the presence of CKD at baseline in a multivariate analysis adjusted for the significant variables identified in univariate analysis (Table 2). 3.2. Follow-up assessments Of the 7885 participants without CKD at baseline, 1824 (23.1%) had 5-year follow-up assessments. In comparison of participants with and without follow-up assessment, no significant differences were found in baseline age, sex, habitual drinking, antidiabetic or antilipidemic medication use, fasting plasma glucose level, HbA1c level, or HDL-cholesterol level; however, significant differences were found in BMI, current smoking, antihypertensive medication use, blood pressure, triglyceride levels, LDL-cholesterol level, eGFR
Table 2 Logistic regression analysis for the presence of CKD at baseline.
Age (years) Body mass index (kg/m2) Smoking habit (vs. absence) Habitual alcohol intake (vs. absence) Medication for hypertension (vs. absence) Medication for diabetes (vs. absence) Medication for dislipidemia (vs. absence) Systolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Carotid IMT (per 0.10 mm)
OR
95% CI
P value
1.08 1.08 0.85 0.86 1.26 0.95 1.42 0.99 1.13 0.66 1.46 0.82 1.06 0.96
1.07e1.09 1.05e1.10 0.66e1.10 0.73e1.02 1.06e1.50 0.61e1.48 1.14e1.78 0.989e0.997 1.06e1.21 0.57e0.77 1.32e1.61 0.66e1.03 0.96e1.16 0.91e1.01
<0.001 <0.001 0.209 0.082 0.009 0.814 0.002 0.001 <0.001 <0.001 <0.001 0.086 0.257 0.127
OR, 95% CI, and P value were calculated by logistic regression analysis. CKD, chronic kidney disease; OR, odds ratio; CI, confidence interval; HDL, high density lipoprotein; LDL, low density lipoprotein; IMT, intima-media thickness.
Table 3 Comparison of participants who did/did not develop CKD during follow-up. Follow-up eGFR
Baseline Age (years) Male (%) Body mass index (kg/m2) Smoking habit (%) Habitual alcohol intake (%) Medication for hypertension (%) Medication for diabetes (%) Medication for dislipidemia (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) eGFR (mL/min/1.73 m2) Carotid IMT (mm) Follow-up Body mass index (kg/m2) Medication for hypertension (%) Medication for diabetes (%) Medication for dislipidemia (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mmol/L) HbA1c (%) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) eGFR (mL/min/1.73 m2)
P value
60 mL/min/ 1.73 m2
<60 mL/min/ 1.73 m2
(N ¼ 1600)
(N ¼ 224)
56.1 ± 10.9 35.1 23.6 ± 3.3 14.1 41.0 14.5
64.5 ± 9.7 34.4 24.8 ± 3.2 7.7 37.4 23.8
2.9 5.9
2.2 10.3
123.8 ± 17.1
130.1 ± 18.3
73.9 ± 11.4
74.8 ± 10.5
0.315
5.4 ± 1.3
5.8 ± 1.9
0.008
5.5 ± 0.6 1.3 ± 0.7 1.6 ± 0.4 3.1 ± 0.8 78.1 ± 11.7 0.65 ± 0.14
5.7 ± 0.8 1.6 ± 0.8 1.5 ± 0.4 3.1 ± 0.9 66.1 ± 5.8 0.74 ± 0.22
0.004 <0.001 <0.001 0.485 <0.001 <0.001
23.3 ± 3.4 23.2
24.4 ± 3.1 39.9
<0.001 <0.001
5.0 12.1
7.6 17.5
124.6 ± 16.6
131.4 ± 17.5
74.4 ± 10.6
74.9 ± 9.2
5.5 ± 1.1
5.9 ± 1.6
<0.001
5.6 ± 0.6 1.3 ± 0.9 1.7 ± 0.4 3.2 ± 0.8 76.3 ± 11.4
5.8 ± 0.7 1.6 ± 1.3 1.5 ± 0.4 3.3 ± 0.8 54.8 ± 5.4
0.003 <0.001 <0.001 0.004 <0.001
<0.001 0.840 <0.001 0.016 0.383 0.001 0.589 0.021 <0.001
0.102 0.025 <0.001 0.476
Data shown as mean ± standard deviation or percentage. P values calculated by unpaired t-test or chi-square test between participants who did/did not develope CKD in the follow-up phase. CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HDL, high density lipoprotein; LDL, low density lipoprotein; IMT, intima-media thickness.
210
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
Table 4 Logistic regression analysis for the development of CKD during follow-up.
Model 1a Baseline Baseline Baseline Model 2b Baseline Baseline Baseline
OR
95% CI
P value
eGFR (mL/min/1.73 m2) Carotid IMT (per 0.10 mm) Triglycerides (mmol/L)
0.83 1.17 1.35
0.80e0.85 1.04e1.33 1.06e1.73
<0.001 0.010 0.015
eGFR (mL/min/1.73 m2) Carotid IMT (per 0.10 mm) Triglycerides (mmol/L)
0.84 1.19 1.41
0.81e0.87 1.05e1.36 1.09e1.83
<0.001 0.008 0.008
OR, 95% CI, and P value were calculated by Logistic regression analysis. CKD, chronic kidney disease; OR, odds ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; IMT, intima-media thickness. a Adjusted for baseline variables age, body mass index, smoking status, medication for hypertension, medication for dyslipidemia, systolic blood pressure, glucose, HbA1c, and HDL-cholesterol. b Adjusted for baseline variables sex, drinking status, medication for diabetes, LDL-cholesterol, and diastolic blood pressure (in addition to the adjustments made for model 1).
Fig. 1. The decrease in eGFR by the tertile of baseline carotid IMT during the five years between baseline and follow-up. Data are shown as adjusted mean values of the change in eGFR and 95% confidence intervals in subgroups classified by tertile of baseline carotid IMT. The change in eGFR was adjusted for the baseline variables age, body mass index, smoking status, medication for hypertension, medication for dyslipidemia, systolic blood pressure, fasting plasma glucose, HbA1c, triglycerides, HDLcholesterol, and eGFR. Intergroup differences were calculated by the Bonferroni test. IMT, intima-media thickness; eGFR, estimated glomerular filtration rate; HDL, high density lipoprotein.
4. Discussion The main finding of the present 5-year prospective cohort study of a Japanese population is that carotid IMT is a significant predictor for the development of CKD and the decline in eGFR, independently of the traditional risk factors. Moreover, triglycerides were also significantly associated with the development of CKD. These results suggest that the evaluation of carotid atherosclerosis and serum triglyceride levels may be useful for the early detection of CKD in relatively healthy individuals without a history of cardiovascular disease. In this analysis, carotid IMT was an independent predictor for the development of CKD; while the traditional risk factors of age, BMI, systolic blood pressure, fasting plasma glucose, and HbA1c were not related in multivariate analysis that included carotid IMT. Chonchol et al. previously reported that carotid IMT, systolic blood
pressure, HbA1c, and baseline eGFR were significant determinants of the development of CKD [12]. We believe that the different characteristics of the population we studied, in which healthy participants were the majority, may be responsible for these discrepancies. The population of Chonchol's study was aged >55 years and included patients with a history of cardiovascular disease (about 10%). The mean levels of systolic blood pressure (139 mmHg), HbA1c (5.8%), and carotid IMT (0.78 mm) were also higher in their study. Therefore, we feel that the impact of blood pressure and HbA1c may be relatively low and that carotid IMT would be a superior indicator of the future development of CKD for people with low cardiovascular risk. The presence of carotid atherosclerosis could indicate nephrosclerosis. A positive relationship between carotid IMT and nephrosclerosis was reported among kidney transplant donors [23]. Thus, for relatively healthy people, traditional risk factors may induce the decline of kidney function through the progression of nephrosclerosis. This is a possible explanation as to why carotid atherosclerosis is useful for predicting future CKD and the decline in eGFR, over and above the traditional risk factors. Carotid IMT was not significantly correlated with the presence of CKD in the baseline multivariate analysis of this study. Some previous cross-sectional studies have shown a significant relationship between carotid IMT and eGFR in patients with hypertension [24,25], with impaired glucose metabolism [24,26], or with coronary artery disease [27]. In contrast, among relatively healthy subjects, this relationship was not found in univariate [28,29] or multivariate analyses [30e32], which is similar to our results. The significant cross-sectional correlation between carotid IMT and eGFR may be limited to subjects with advanced atherosclerosis and/or kidney dysfunction. In the present study, the baseline triglyceride levels were positively associated with the development of CKD, independently of carotid IMT. A few community-based studies have shown a correlation between hypertriglyceridemia and kidney dysfunction [33,34]. Furthermore, among patients with type 2 diabetes, a relationship between hypertriglyceridemia and diabetic kidney disease with eGFR < 60 mL/min/1.73 m2 and/or microalbuminuria was shown by an international case-control study [35]. Hypertriglyceridemia is related to insulin resistance [36], which has been reported to be a risk factor for the decline of kidney function [37]. Renal accumulation of triglycerides and free fatty acids has also been suggested to be associated with kidney injury and dysfunction [38]. This, taken together with our findings, indicates that hypertriglyceridemia contributes to the development of CKD through insulin resistance (and compensatory hyperinsulinemia) or lipotoxity, independently of atherosclerosis. Several limitations of this study bear mention. First, the possibility of selection bias because of the exclusion of the participants who were not available for follow-up assessment needs to be considered when generalizing the present findings. Follow-up assessments were done for only 23.1%. Second, the diagnosis of CKD was based on a single examination, which could lead to misclassification. Third, data regarding microalbuminuria and proteinuria were not collected, and no information about the etiology of the CKD was obtained. Fourth, non-traditional risk factors associated with atherosclerosis or kidney dysfunction, such as inflammatory markers, insulin and insulin resistance markers, and homocysteine, were not explored in the present study. Finally, we could not evaluate the relationship between the 5-year change in carotid IMT and that of eGFR. The impact of both time-related and treatmentinduced changes of carotid atherosclerosis on the decline of kidney function should be evaluated in future research. Despite the above limitations, we believe that our findings from this large community-based sample will contribute the prevention of CKD in
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
healthy individuals. 5. Conclusions We found that subclinical carotid atherosclerosis is predictive of the future incidence of CKD in a Japanese general population. Moreover, serum triglyceride levels were also independently associated with the development of CKD. Further research should focus on which groups in the general population would benefit most from the evaluation of carotid atherosclerosis. In addition, more research is needed to investigate whether or not aggressive treatment for cardiovascular risk factors among persons with subclinical atherosclerosis can improve renal prognosis thorough the prevention or reduction of atherosclerosis. Financial support This study was supported by the Japan Multi-institutional Collaborative Cohort Study (J-MICC Study), a Scientific Support Program for Cancer Research Grant-in-Aid for Scientific Research on Innovative Areas (No. 221S001), and a Grant-in-Aid for Comprehensive Research of the 21st Century COE Program from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Conflict of interest This paper has not been published or presented elsewhere in part or in its entirety and is not under consideration by another journal. M.S, F.M, K.T, K.U, S.H, H.I, T.I, H.M, E.O, K.T, M.K, M.M, and J.H. have nothing to declare. N.F. has received support from Daiichi Sankyo Healthcare Co., Ltd., Tokyo, Japan, Chugai Pharmaceutical Co., Ltd., Tokyo, Japan, MSD Ltd, Tokyo, Japan and Mitsubishi Tanabe Pharma, Osaka, Japan. The remaining authors disclose no conflicts. Acknowledgments We are grateful to Drs. Kyoko Okada, Hiroaki Taniai, Hachiro Ohnishi, Tsunehisa Koga, Kunimitsu Eiraku, Takeo Hayashi, Yuji Harada, Sakiko Hayasaki, Ayaka Komori, Rinne Sakemi, Sho Yamasaki, Azusa Hatashima, Yoshifumi Kato, Yuki Tanaka, and Masaru Sakiyama from our department for their assistance. We also thank Ms. Setsuko Nagata and Takako Nagamine for their invaluable assistance. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2014.12.013. References [1] S.I. Hallan, J. Coresh, B.C. Astor, et al., International comparison of the relationship of chronic kidney disease prevalence and ESRD risk, J. Am. Soc. Nephrol. 17 (2006) 2275e2284. [2] J. Coresh, E. Selvin, L.A. Stevens, et al., Prevalence of chronic kidney disease in the United States, J. Am. Med. Assoc. 298 (2007) 2038e2047. [3] E. Imai, M. Horio, T. Watanabe, et al., Prevalence of chronic kidney disease in the Japanese general population, Clin. Exp. Nephrol. 13 (2009) 621e630. [4] P. Muntner, J. He, L. Hamm, et al., Renal insufficiency and subsequent death resulting from cardiovascular disease in the United States, J. Am. Soc. Nephrol. 13 (2002) 745e753. [5] C. Meisinger, A. Doring, H. Lowel, Chronic kidney disease and risk of incident myocardial infarction and all-cause and cardiovascular disease mortality in middle-aged men and women from the general population, Eur. Heart J. 27 (2006) 1245e1250. [6] N.B. Shulman, C.E. Ford, W.D. Hall, et al., Prognostic value of serum creatinine and effect of treatment of hypertension on renal function. Results from the hypertension detection and follow-up program. The Hypertension Detection
211
and Follow-up Program Cooperative Group, Hypertension 13 (1989) I80eI93. [7] K. Yamagata, K. Ishida, T. Sairenchi, et al., Risk factors for chronic kidney disease in a community-based population: a 10-year follow-up study, Kidney Int. 71 (2007) 159e166. [8] M. Tonelli, C. Wanner, Lipid management in chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2013 clinical practice guideline, Ann. Intern. Med. 160 (2014) 182e189. [9] M. Shimizu, J. Ishikawa, K. Eguchi, et al., Association of an abnormal blood glucose level and morning blood pressure surge in elderly subjects with hypertension, Am. J. Hypertens. 22 (2009) 611e616. [10] A. Cases, E. Coll, Dyslipidemia and the progression of renal disease in chronic renal failure patients, Kidney Int. Suppl. (2005) S87eS93. [11] M.W. Lorenz, H.S. Markus, M.L. Bots, et al., Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and metaanalysis, Circulation 115 (2007) 459e467. [12] M. Chonchol, H. Gnahn, D. Sander, Impact of subclinical carotid atherosclerosis on incident chronic kidney disease in the elderly, Nephrol. Dial. Transpl. 23 (2008) 2593e2598. [13] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499e502. [14] H. Ohnishi, Y. Sawayama, N. Furusyo, et al., Risk factors for and the prevalence of peripheral arterial disease and its relationship to carotid atherosclerosis: the Kyushu and Okinawa Population Study (KOPS), J. Atheroscler. Thromb. 17 (2010) 751e758. [15] N. Furusyo, T. Koga, M. Ai, et al., Utility of glycated albumin for the diagnosis of diabetes mellitus in a Japanese population study: results from the Kyushu and Okinawa Population Study (KOPS), Diabetologia 54 (2011) 3028e3036. [16] M. Unno, N. Furusyo, H. Mukae, et al., The utility of visceral fat level by bioelectrical impedance analysis in the screening of metabolic syndrome - the results of the Kyushu and Okinawa Population Study (KOPS), J. Atheroscler. Thromb. 19 (2012) 462e470. [17] N. Furusyo, T. Koga, M. Ai, et al., Plasma glycated albumin level and atherosclerosis: results from the Kyushu and Okinawa Population Study (KOPS), Int. J. Cardiol. 167 (2013) 2066e2072. [18] T. Ihara, N. Furusyo, T. Hayashi, et al., A population-based epidemiological survey of human parvovirus B19 infection: a project of the Kyushu and Okinawa Population Study (KOPS), Arch. Virol. 158 (2013) 2465e2472. [19] H. Ikezaki, N. Furusyo, K. Okada, et al., The utility of urinary myo-inositol as a marker of glucose intolerance, Diabetes Res. Clin. Pract. 103 (2014) 88e96. [20] N. Furusyo, M. Ai, M. Okazaki, et al., Serum cholesterol and triglyceride reference ranges of twenty lipoprotein subclasses for healthy Japanese men and women, Atherosclerosis 231 (2013) 238e245. [21] A. Coleman, S. Steel, P. Freeman, et al., Validation of the Omron M7 (HEM-780E) oscillometric blood pressure monitoring device according to the British Hypertension Society protocol, Blood Press Monit. 13 (2008) 49e54. [22] S. Matsuo, E. Imai, M. Horio, et al., Revised equations for estimated GFR from serum creatinine in Japan, Am. J. Kidney Dis. 53 (2009) 982e992. [23] S. Erten, O. Gungor, S. Sen, et al., Nephrosclerosis and carotid atherosclerosis: lessons from kidney donor histology, Nephrol. (Carlton) 16 (2011) 720e724. [24] N. Ishizaka, Y. Ishizaka, E. Toda, et al., Association between chronic kidney disease and carotid intima-media thickening in individuals with hypertension and impaired glucose metabolism, Hypertens. Res. 30 (2007) 1035e1041. [25] T. Ohara, Y. Kokubo, K. Toyoda, et al., Impact of chronic kidney disease on carotid atherosclerosis according to blood pressure category: the Suita study, Stroke 44 (2013) 3537e3539. [26] H. Ito, Y. Komatsu, M. Mifune, et al., The estimated GFR, but not the stage of diabetic nephropathy graded by the urinary albumin excretion, is associated with the carotid intima-media thickness in patients with type 2 diabetes mellitus: a cross-sectional study, Cardiovasc Diabetol. 9 (2010) 18. [27] A. Lisowska, W.J. Musial, P. Lisowski, et al., Intima-media thickness is a useful marker of the extent of coronary artery disease in patients with impaired renal function, Atherosclerosis 202 (2009) 470e475. [28] P. Dasmahapatra, S.R. Srinivasan, J. Mokha, et al., Subclinical atherosclerotic changes related to chronic kidney disease in asymptomatic black and white young adults: the Bogalusa heart study, Ann. Epidemiol. 21 (2011) 311e317. [29] X. Kong, X. Jia, Y. Wei, et al., Association between microalbuminuria and subclinical atherosclerosis evaluated by carotid artery intima-media in elderly patients with normal renal function, BMC Nephrol. 13 (2012) 37. [30] L. Zhang, F. Zhao, Y. Yang, et al., Association between carotid artery intimamedia thickness and early-stage CKD in a Chinese population, Am. J. Kidney Dis. 49 (2007) 786e792. [31] S.W. Choi, H.Y. Kim, Y.H. Lee, et al., eGFR is associated with subclinical atherosclerosis independent of albuminuria: the Dong-gu Study, Atherosclerosis 212 (2010) 661e667. [32] L. Han, X. Bai, H. Lin, et al., Lack of independent relationship between agerelated kidney function decline and carotid intima-media thickness in a healthy Chinese population, Nephrol. Dial. Transpl. 25 (2010) 1859e1865. [33] P. Muntner, J. Coresh, J.C. Smith, et al., Plasma lipids and risk of developing renal dysfunction: the atherosclerosis risk in communities study, Kidney Int. 58 (2000) 293e301. [34] F. Wang, P. Ye, L. Luo, et al., Association of risk factors for cardiovascular disease and glomerular filtration rate: a community-based study of 4,925 adults in Beijing, Nephrol. Dial. Transpl. 25 (2010) 3924e3931. [35] F.M. Sacks, M.P. Hermans, P. Fioretto, et al., Association between plasma
212
M. Shimizu et al. / Atherosclerosis 238 (2015) 207e212
triglycerides and high-density lipoprotein cholesterol and microvascular kidney disease and retinopathy in type 2 diabetes mellitus: a global casecontrol study in 13 countries, Circulation 129 (2014) 999e1008. [36] T. McLaughlin, F. Abbasi, K. Cheal, et al., Use of metabolic markers to identify overweight individuals who are insulin resistant, Ann. Intern Med. 139 (2003) 802e809.
[37] G. Dogra, A. Irish, D. Chan, et al., Insulin resistance, inflammation, and blood pressure determine vascular dysfunction in CKD, Am. J. Kidney Dis. 48 (2006) 926e934. [38] I.A. Bobulescu, Renal lipid metabolism and lipotoxicity, Curr. Opin. Nephrol. Hypertens. 19 (2010) 393e402.