Circulating resistin is increased with decreasing renal function in a general Japanese population: the Hisayama Study

Abstract

Background. The purpose of this study is to investigate the relationship between serum resistin levels and chronic kidney disease (CKD).

Methods. A total of 3192 community-dwelling subjects (1377 men, 1815 women), aged ≥40 years and without renal failure, were divided into four groups according to quartiles of serum resistin concentrations: ≤7.1, 7.2–9.9, 10.0–14.7 and ≥14.8 ng/mL. The associations of resistin levels with renal function status were examined cross-sectionally. The estimated glomerular filtration rate (eGFR) was calculated using the equation from the Modification of Diet in Renal Disease Study, and CKD was defined as an eGFR of <60 mL/min/1.73 m².

Results. The age- and sex-adjusted mean values of eGFR decreased significantly with elevating quartiles of resistin (P for trend <0.001). The age- and sex-adjusted odds ratios (ORs) for the presence of CKD increased progressively with higher quartiles of resistin. This trend remained robust even after controlling for age, sex, body mass index, diabetes, homeostasis model assessment of insulin resistance (HOMA-IR), high-sensitivity C-reactive protein (hs-CRP), triglycerides, high-density lipoprotein and total cholesterol, hypertension, current smoking, current drinking, and regular exercise [second quartile: OR 1.44, 95% confidence interval (CI) 1.05–1.99; third quartile: OR 2.15, 95% CI 1.58–2.92; fourth quartile: OR 2.32, 95% CI 1.71–3.16; P for trend <0.001]. In stratified analyses, high resistin level (≥7.2 ng/mL) was a significant relevant factor in CKD, independent of HOMA-IR or hs-CRP level.

Conclusion. Our findings suggest that elevated resistin level is significantly associated with the likelihood of CKD in the general Japanese population.

chronic kidney disease, cross-sectional study, epidemiology, resistin

Introduction

Chronic kidney disease (CKD) is a worldwide public health concern and a major risk factor for end-stage renal disease, cardiovascular disease and premature death [1]. Identifying and treating risk factors for mild CKD may be the best approach to prevent and delay advanced outcomes [1]. Several epidemiological studies associated age, high blood pressure, diabetes, proteinuria, dyslipidaemia and smoking with the subsequent decline in estimated glomerular filtration rate (eGFR) [2,3]. It was also reported that insulin resistance and inflammation were emerging risk factors for the occurrence of CKD [4,5]. However, regardless of the treatment and prevention of these factors, patients with renal failure are increasing in number [6], suggesting the presence of other risk factors.

Resistin belongs to a family of cysteine-rich secretory proteins called resistin-like molecules [7]. In rodents, resistin is derived almost exclusively from fat tissue, and its serum levels are elevated in animal models of obesity and insulin resistance [8]. In humans, on the other hand, resistin is highly expressed in monocytes and macrophages [9]; thus, its pathophysiological role may differ between species. In vitro, resistin activated human endothelial cells, leading to increased expression of adhesion molecules, and induced human aortic muscle cell proliferation [10]. Furthermore, some clinical and epidemiological studies revealed positive correlations between plasma resistin levels and pro-inflammatory cytokines [11,12]. A few recent clinical studies also showed an inverse correlation between resistin level and eGFR in CKD patients [13–15]. To date, however, there have been no investigations into the link between serum resistin levels and CKD in large general populations excluding patients with renal failure. The aim of the present study is to examine the relationship between serum resistin levels and CKD in a cross-sectional study of a general Japanese population.

Materials and methods

Study population

A population-based prospective study of cardiovascular disease has been under way since 1961 in the town of Hisayama, a suburb of Fukuoka City in southern Japan. As part of this study, in 2002, we conducted a cross-sectional examination among residents of the town. A detailed description of this survey was published previously [16]. Briefly, of all residents aged 40 years or over, 3328 underwent the examination (participation rate, 77.6%). After excluding 30 subjects who did not consent to participate in the study, 13 subjects with renal failure (eGFR <15 mL/min/1.73 m² or treated by dialysis), 82 subjects who had already eaten breakfast on the day serum samples were to be taken and 11 subjects without serum samples for resistin measurement, the remaining 3192 subjects (1377 men, 1815 women) were enrolled in this study.

This study was conducted with the approval of the Ethics Committee of the Faculty of Medicine, Kyushu University. Written informed consent was obtained from all participants.

Clinical evaluation and laboratory measurements

At the screening examination, blood samples were collected from an antecubital vein between 8:00 and 10:30 a.m. after at least a 12-h overnight fast. A portion of each serum specimen was stored at −80°C for 5 years, until 2007, when it was used for the measurement of resistin concentrations by a human resistin enzyme-linked immunosorbent assay kit supplied by R&D Systems (Minneapolis, MN) following the manufacturer’s protocol. Linearity was maintained <0.16 ng/mL, and both intra- and inter-assay coefficient variations were comparable to those specified by the manufacturer (2.6–10.5%). Using fresh blood samples, serum creatinine was measured by the enzymatic method. Levels of triglycerides, high-density lipoprotein (HDL) and total cholesterol were determined enzymatically. Blood for the glucose assay was obtained by venipuncture into tubes containing sodium fluoride, and plasma glucose concentrations were determined by the glucose oxidase method. Subjects were considered to have diabetes mellitus if they had a fasting plasma glucose level of ≥7.0 mmol/L, had a 2-h post-load glucose level of ≥11.1 mmol/L or were taking antidiabetic medications. Serum insulin values were measured by a chemiluminescent enzyme immunoassay. The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated with the formula fasting plasma glucose (mmol/L) × fasting serum insulin (μU/mL) / 22.5 [17], and subjects in the top quartile of HOMA-IR distribution were defined as having insulin resistance [18]. High-sensitivity C-reactive protein (hs-CRP) levels were quantified using a modification of the Behring latex-enhanced CRP assay on a Behring nephelometer BN-100 (Behring Diagnostics, Westwood, MA). High CRP values were defined as ≥1.0 mg/L, according to our previous report [19].

Blood pressure was obtained three times using an automated sphygmomanometer (BP-203RV III; Colin, Tokyo, Japan) with the subjects in a sitting position; the average of the three measurements was used in the present analysis. Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg or current treatment with antihypertensive agents. Height and weight were measured with the subject wearing light clothes without shoes, and body mass index [BMI (kg/m²)] was calculated.

Each participant completed a self-administered questionnaire covering medical history, smoking habit, alcohol intake and exercise. The questionnaire was checked by trained interviewers at the screening. Smoking habit and alcohol intake were classified as either current habitual use or not. Those subjects who engaged in sports or other forms of exertion ≥3 times a week during their leisure time made up a regular exercise group.

Definition of CKD

GFR was estimated by using the following modified equation of the Modification of Diet in Renal Disease (MDRD) Study for Japanese [20]: eGFR (mL/min/1.73 m²) = 175 × [serum creatinine (mg/dL)]^−1.154 × [age (years)]^−0.203 × [0.741 (Japanese coefficient)] × (0.742 if female). Based on the National Kidney Foundation Kidney Disease Outcomes Quality Initiative Clinical Practice Guidelines [1], we divided kidney function levels into four categories according to eGFR: normal and CKD stage 1 (eGFR ≥90 mL/min/1.73 m²), CKD stage 2 (eGFR 60–89 mL/min/1.73 m²), CKD stage 3 (eGFR 30–59 mL/min/1.73 m²) and CKD stage 4 (eGFR 15–29 mL/min/1.73 m²). We also determined CKD as a dichotomized category when eGFR was <60 mL/min/1.73 m².

Statistical analysis

SAS software package version 8.2 (SAS Institute, Cary, NC) was used to perform all statistical analyses. Because the distributions of fasting insulin, HOMA-IR, hs-CRP and triglycerides were skewed, these variables were natural log-transformed for statistical analysis. The subjects were divided into quartiles of resistin concentrations: ≤7.1, 7.2–9.9, 10.0–14.7 and ≥14.8 ng/mL. The mean values of possible risk factors were adjusted for age and sex using the analysis of covariance and were compared among the resistin quartiles according to the linear regression model. Age- and sex-adjusted means of eGFR among the quartiles were determined by the same method. The frequencies of risk factors were adjusted for age and sex by the direct method using all subjects as a standard population and were tested for trends using logistic regression analysis. The age- and sex- or multivariate-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for CKD were also determined by logistic regression analysis.

Results

Table 1 shows the age- and sex-adjusted means or frequencies of potential risk factors according to the quartiles of serum resistin concentrations. The mean values of age, serum creatinine, fasting insulin, HOMA-IR and hs-CRP, as well as the percentage of men, increased with the quartiles of resistin values, while the mean HDL cholesterol concentration and the frequencies of alcohol intake and regular exercise were negatively correlated with the quartiles. The other variables were not significantly associated with the quartiles.

Table 1

Age- and sex-adjusted mean values or frequencies of risk factors according to quartiles of serum resistin concentrations

Variable	Serum resistin levels (ng/mL)				P-value for trend
	1.5–7.1	7.2–9.9	10.0–14.7	14.8–90.2
	(n = 788)	(n = 803)	(n = 805)	(n = 796)
Age (years)	59 ± 11	61 ± 12	62 ± 13	64 ± 13	<0.001
Men (%)	38.1	40.4	44.6	49.5	<0.001
BMI (kg/m2)	23.0 ± 3.4	23.1 ± 3.4	23.0 ± 3.4	23.2 ± 3.4	0.50
Serum creatinine (μmol/L)	59 ± 15	62 ± 15	64 ± 15	65 ± 15	<0.001
Fasting plasma glucose (mmol/L)	6.1 ± 1.3	6.1 ± 1.3	6.0 ± 1.3	6.1 ± 1.3	0.70
Fasting insulin (pmol/L)	44.6 (13.8–144.5)	46.6 (14.5–150.0)	46.3 (14.4–148.7)	49.5 (15.3–160.5)	0.001
Diabetes (%)	16.5	16.8	18.0	18.9	0.45
HOMA-IR	1.66 (0.44–6.29)	1.72 (0.46–6.48)	1.71 (0.45–6.42)	1.83 (0.48–6.94)	0.007
Hs-CRP (mg/L)	0.44 (0.04–4.53)	0.51 (0.05–5.28)	0.54 (0.05–5.49)	0.66 (0.06–6.88)	<0.001
Triglycerides (mmol/L)	1.12 (0.39–3.18)	1.16 (0.41–3.29)	1.14 (0.40–3.23)	1.17 (0.41–3.35)	0.13
HDL-cholesterol (mmol/L)	1.70 ± 0.40	1.61 ± 0.40	1.62 ± 0.40	1.55 ± 0.40	<0.001
Total cholesterol (mmol/L)	5.30 ± 0.90	5.27 ± 0.89	5.31 ± 0.89	5.23 ± 0.90	0.15
Systolic blood pressure (mmHg)	132 ± 20	132 ± 20	132 ± 20	132 ± 20	0.96
Diastolic blood pressure (mmHg)	79 ± 12	79 ± 12	78 ± 12	78 ± 12	0.66
Hypertension (%)	45.0	42.9	43.2	46.3	0.98
Current smoking (%)	20.7	21.5	21.9	22.8	0.29
Current drinking (%)	48.8	44.4	43.4	38.3	<0.001
Regular exercise (%)	11.1	10.4	9.8	8.7	0.04

Values are given as means ± SD or frequencies. Age and percentage of men are not adjusted. Fasting insulin, HOMA-IR, hs-CRP and triglycerides are shown by geometric means and 95% CIs due to the skewed distribution.

BMI, body mass index; HOMA-IR, homeostasis model assessment of insulin resistance; hs-CRP, high-sensitivity C-reactive protein; HDL, high-density lipoprotein.

The left panel in Figure 1 shows the age- and sex-adjusted mean values of eGFR according to the quartiles of serum resistin values. The mean values of eGFR decreased significantly with the quartiles of resistin values (80.4, 76.7, 74.2 and 73.9 mL/min/1.73 m², respectively; P for trend <0.001). Meanwhile, the age- and sex-adjusted mean values of serum resistin according to eGFR are shown in the right panel in Figure 1. The age- and sex-adjusted geometric mean values of serum resistin increased significantly as eGFR decreased (9.5 ng/mL in the normal and CKD stage 1, 10.2 ng/mL in CKD stage 2, 11.8 ng/mL in CKD stage 3 and 21.1 ng/mL in CKD stage 4; P for trend <0.001): the differences were significant between normal and CKD stage 1 and CKD stages 2–4 (all P < 0.001).

Fig. 1

The age- and sex-adjusted mean values of estimated glomerular filtration rate (eGFR) according to quartiles of serum resistin concentrations (left panel), and the age- and sex-adjusted mean values of serum resistin levels according to eGFR (right panel). Values are given as means ± standard error. *P < 0.001 vs the first quartile, †P < 0.001 vs eGFR of ≥90 mL/min/1.73 m².

Table 2 shows the age- and sex-adjusted or multivariate-adjusted ORs and 95% CIs for the presence of CKD according to the resistin quartiles. The age- and sex-adjusted OR for CKD significantly increased with elevating quartiles (P for trend <0.001); compared to the first quartile, OR was greater in the second to fourth quartiles. Such associations were substantially unchanged after adjustment for age, sex, BMI, diabetes, HOMA-IR, hs-CRP, triglycerides, HDL and total cholesterol, hypertension, current smoking, current drinking, and regular exercise (second quartile: OR 1.44, 95% CI 1.05–1.99; third quartile: OR 2.15, 95% CI 1.58–2.92; fourth quartile: OR 2.32, 95% CI 1.71–3.16).

Table 2

Age- and sex- or multivariate-adjusted ORs and their 95% CIs for the presence of chronic kidney disease according to quartiles of serum resistin concentrations

	Serum resistin levels (ng/mL)				P-value for trend
	1.5–7.1	7.2–9.9	10.0–14.7	14.8–90.2
Subjects (n)	788	803	805	796
CKD cases (n)	75	117	166	192
Age- and sex-adjusted OR (95% CI)	1 (reference)	1.44 (1.06–1.98)	2.15 (1.59–2.90)	2.33 (1.73–3.14)	<0.001
Multivariate-adjusted OR (95% CI)	1 (reference)	1.44 (1.05–1.99)	2.15 (1.58–2.92)	2.32 (1.71–3.16)	<0.001

Multivariate adjustment was made for age, sex, body mass index, diabetes, homeostasis model assessment of insulin resistance, high-sensitivity C-reactive protein, triglycerides, high-density lipoprotein and total cholesterol, hypertension, current smoking, current drinking, and regular exercise.

CKD, chronic kidney disease; OR, odds ratio; CI, confidence interval.

Finally, we examined the combined as well as separate effects of resistin and HOMA-IR or hs-CRP levels on the presence of CKD (Table 3). When a high resistin level was in the second or higher quartile (≥7.2 ng/mL), the age- and sex-adjusted ORs of CKD were significantly higher in subjects with high resistin and low HOMA-IR (<75th percentile) and in subjects with high resistin and high HOMA-IR (≥75th percentile) compared to the reference group, who had low resistin and low HOMA-IR. Similarly, the age- and sex-adjusted risks of CKD were significantly higher in subjects with high resistin levels, independent of hs-CRP levels. These relationships remained robust even after adjusting for the confounding factors named above.

Table 3

Age- and sex- or multivariate-adjusted ORs and their 95% CIs for chronic kidney disease according to the presence or absence of high resistin levels and high HOMA-IR as well as high hs-CRP values

	Subjects, n	CKD cases, n	Age- and sex-adjusted OR (95% CI)	P-value	Multivariate-adjusted OR (95% CI)	P-value
HOMA-IR
Low + low resistin	602	57	1 (reference)		1 (reference)
Low + high resistin	1773	340	1.84 (1.35–2.49)	<0.001	1.85 (1.35–2.52)	<0.001
High + low resistin	178	17	1.00 (0.56–1.78)	0.99	0.93 (0.51–1.68)	0.80
High + high resistin	613	131	2.34 (1.66–3.29)	<0.001	2.14 (1.47–3.12)	<0.001
hs-CRP
Low + low resistin	645	53	1 (reference)		1 (reference)
Low + high resistin	1717	309	2.13 (1.56–2.91)	<0.001	2.12 (1.54–2.91)	<0.001
High + low resistin	143	22	1.69 (0.98–2.92)	0.06	1.62 (0.92–2.84)	0.09
High + high resistin	687	166	2.45 (1.74–3.45)	<0.001	2.46 (1.72–3.50)	<0.001

High resistin levels were defined as the second or higher quartiles of its values; low resistin levels were the first quartile of its values.

HOMA-IR: ‘high’ indicates ≥75th percentile (HOMA-IR ≥2.6); ‘low’ <75th percentile.

Hs-CRP: ‘high’ indicates ≥1.0 mg/L; ‘low’ <1.0 mg/L.

Multivariate adjustment was made for age, sex, body mass index, diabetes, HOMA-IR, hs-CRP, triglycerides, high-density lipoprotein and total cholesterol, hypertension, current smoking, current drinking, and regular exercise, but each risk factor that had been used for categorization was excluded from the confounding factors.

CKD, chronic kidney disease; HOMA-IR, homeostasis model assessment of insulin resistance; hs-CRP, high-sensitivity C-reactive protein; OR, odds ratio; CI, confidence interval.

Discussion

Using the large cross-sectional data of a general Japanese population, we demonstrated that serum resistin levels were negatively associated with eGFR, and that the mean values of serum resistin increased even in CKD stage 2. Furthermore, the elevated levels of serum resistin were an independent relevant factor for CKD after controlling for age, sex, BMI, diabetes, HOMA-IR, hs-CRP, triglycerides, HDL and total cholesterol, hypertension, current smoking, current drinking, and regular exercise. In stratified analyses, high resistin levels were associated significantly with the likelihood of CKD in those with low HOMA-IR (<75th percentile) as well as in those with low hs-CRP levels (<1.0 mg/L). These findings suggest that the measurement of resistin values provides additional information on the risk factors for CKD.

In a few clinical studies of CKD patients, resistin levels have been shown to be inversely correlated with eGFR [13–15]. Although polypeptides that have molecular weights comparable to those of resistin are thought to be freely filtered at the normal glomerulus [21], subjects with advanced renal impairment might have serum resistin accumulations due to reduced renal clearance: that is, renal dysfunction might cause elevated serum resistin levels. However, the present study showed that resistin levels were significantly raised even in subjects with CKD stage 2 (eGFR of 60–89 mL/min/1.73 m²), in which polypeptides would be filtered almost normally. In a clinical study of patients with immunoglobulin A glomerulonephritis, serum resistin levels were also significantly higher in subjects with mild renal dysfunction who had a mean GFR of 76 mL/min/1.73 m² than in those who had a mean GFR of 114 mL/min/1.73 m² [13]. These observations support the hypothesis that resistin potentially plays an important role in the development of CKD.

It has been assumed that the effect of resistin is mediated via insulin resistance or inflammation. In our study, serum resistin levels were positively associated with HOMA-IR, but the association between circulating resistin levels and the likelihood of CKD was independent of HOMA-IR, which is in accord with the results of other clinical studies [13,14,22]. Thus, insulin resistance may not be a major mediator of the association between resistin levels and the risk of CKD. Meanwhile, the present study also indicated that the association between resistin levels and the likelihood of CKD was unrelated to hs-CRP, though resistin levels were closely associated with those levels. Resistin directly stimulated the expression of pro-inflammatory cytokines such as tumour necrosis factor-α and interleukin-6 in human peripheral blood mononuclear cells [23], and increased the expression of vascular cell adhesion molecule-1 and intracellular adhesion molecule-1 genes in vascular endothelial cells [10]. Furthermore, a recent clinical study provided evidence that the association between plasma resistin levels and plasma monocyte chemoattractant protein-1 concentrations was independent of hs-CRP levels [24]. Since it has become apparent that CKD is a state of chronic glomerular inflammation [5], resistin may play an important role in potentiating inflammation in the kidney, irrespective of hs-CRP and other confounding factors.

There are a variety of methods for estimating GFR. Although inulin clearance is the gold standard among these methods, it is difficult to perform in routine practice. During the past decade, a great deal of effort has been devoted to establishing an equation for estimating GFR. There is a debate about which equation is optimal for Japanese; the original MDRD Study equation is widely accepted by clinical practitioners, but it may be unsuitable for Asian populations [20]. When some existing equations, such as the original MDRD Study equations [25] or the Cockcroft–Gault formula [26], were used instead of the modified MDRD Study equation for Japanese in our subjects, we also found significant associations between elevated resistin levels and CKD (data not shown). These findings imply a robust association between serum resistin levels and CKD.

A limitation of our study should be discussed. Due to the study’s cross-sectional design, we cannot exclude the possibility that hyperresistinaemia is a consequence of CKD, which is a condition of low urine excretion from serum. However, we found significantly increased resistin levels even when eGFR was 60–89 mL/min/1.73 m². Thus, we believe that elevated resistin levels are a potential risk factor for the development of CKD.

In conclusion, an elevated resistin level was a significant relevant factor for CKD in a general Japanese population after taking into account other risk factors, including HOMA-IR and hs-CRP. Because of the cross-sectional design of this study, it is still unclear whether or not hyperresistinaemia is a cause of CKD. Further prospective studies are needed to clarify the causative relationship between serum resistin concentrations and CKD.

This study was supported in part by Grants-in-Aid for Scientific Research A (no. 18209024) and C (no. 20591063) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and a Health and Labour Sciences Research Grant of the Ministry of Health, Labour and Welfare of Japan (Comprehensive Research on Aging and Health: H20-Chouju-004). The authors thank the staff of the Division of Health and Welfare of Hisayama for their cooperation in this study.

Conflict of interest statement. None declared.

© Oxford University Press 2010