TY - JOUR
AU - Tai, E. Shyong
AB - Abstract Background. C-reactive protein (CRP), a marker for inflammation, has been recently associated with early kidney damage. We examined the association between CRP and micro/macroalbuminuria in a multi-ethnic Asian population using data from two population-based studies in Singapore. Methods. We studied 5127 individuals, who participated in two separate, cross-sectional studies, the Singapore Prospective Study Program [SP2]/Singapore Cardiovascular Cohort Study 2 (SCCS2), involving 4233 participants of Chinese, Malay and Indian ethnicity, aged 24–95 years in Singapore, and the Singapore Malay Eye Study (SiMES), involving 894 participants of Malay ethnicity, aged 40–80 years. Micro/macroalbuminuria was defined as urinary albumin-to-creatinine ratio of ≥17 mg/g for men and ≥25 g/g for women. CRP was analyzed as a continuous variable and as categories (<1, 1–3, >3 mg/L). Results. The prevalence of micro/macroalbuminuria in the whole population was 21.1%. The prevalence increased with increasing categories of CRP. Compared with persons with CRP concentrations <1 mg/L, the multivariable odds ratio (OR) (95% confidence interval [CI]) was 1.33 (1.11–1.60) in persons with CRP concentrations 1–3 mg/L and 1.60 (1.30–1.96) in persons with CRP concentrations >3 mg/L; P trend <0.0001. In continuous analysis, each unit increase in log CRP was associated with an OR (95% CI) of 1.20 (1.11–1.28) of having micro/macroalbuminuria (P < 0.0001). This association was independent of potential confounders and was consistent across the two study cohorts with similar effect estimates (OR = 1.6) for micro/macroalbuminuria. Conclusions. Elevated CRP levels are associated with micro/macroalbuminuria independent of diabetes, hypertension and other potential confounders. This suggests that inflammation may play a role in early kidney damage. albuminuria, inflammation, kidney damage, microvascular disease, Singapore Introduction Inflammation plays a central role in the pathogenesis of renal [1,2] and cardiovascular disease [3,4]. Microalbuminuria, a marker for early kidney damage [5], is an established risk factor for progressive kidney disease [6,7] and cardiovascular morbidity [8] and mortality [9]. Epidemiological studies have shown that C-reactive protein (CRP), the most extensively studied inflammatory marker in clinical studies, is associated with microalbuminuria in diabetic [10,11], nondiabetic [10] and hypertensive persons [12,13]. Recent studies from the USA [14] and Japan [15] have shown that CRP is associated with microalbuminuria in the general population [14]. However, CRP distributions vary substantially between ethnic groups [16]. For example, in Japanese persons, there was a higher prevalence of microalbuminuria at lower levels of CRP as compared to the US population, limiting broad generalizability of findings to all racial/ethnic populations [15]. In the current study, we aimed to study the relationship between CRP and micro/macroalbuminuria in a multi-ethnic Asian population of Chinese, Malays and Indians, using the combined data from two independent, population-based cross-sectional studies in Singapore. Materials and methods Study population We analyzed data from two independent samples, the Singapore Prospective Study Program (SP2)/Singapore Cardiovascular Cohort Study 2 (SCCS2) and the Singapore Malay Eye Study (SiMES). SP2/SCCS2 included participants (n = 7742) from one of four previous cross-sectional studies: Thyroid and Heart Study 1982–1984 [17], National Health Survey 1992 [18], National University of Singapore Heart Study 1993–1995 [19] or National Health Survey 1998 [20]. All studies involved a random sample of individuals from the Singapore population, aged 24 to 95 years, with disproportionate sampling stratified by ethnicity to increase the number of minority ethnic groups (Malays and Asian Indians). From 2003–2007, 10 747 participants were invited to participate by linking their unique national identification numbers with national registries, of which 7742 attended the interview component. Of these 7742 participants, 5157 attended the clinical examination. Detailed population selection and methodology have been previously reported. SiMES is a population-based, cross-sectional study of Malay adults (n = 3280), aged 40–80 years living in Singapore. Of the 4168 eligible participants invited, 3280 participated in the study (78.7% response rate). All participants with diabetes and approximately one in five randomly selected nondiabetic participants had their urine collected for assessment of micro/macroalbuminuria (n = 944). Details of the study participants and methods have been published previously [21]. Participants for both SP2/SCCS2 and SiMES were examined in the same study clinic (Singapore Eye Research Institute), and outcome measurements were identical in both studies. The sample size after combining both the study populations was 11 022. In total, 5481 (4537 in SP2/SCCS2 and 944 in SiMES) provided urine samples for measurement of albumin and creatinine. Of the total 5481 participants, we further excluded those with missing data on CRP (n = 313) and selected other variables (n = 41), leaving 5127 for the current analysis. Compared with those included in the final analysis, excluded participants were older; Malays and smokers had higher prevalence of diabetes, had higher levels of CRP, body mass index (BMI), systolic blood pressure (BP) and total cholesterol; lower levels of HDL cholesterol; and less likely to be drinkers. Both studies followed the Declaration of Helsinki and were approved by the local Institutional Review Boards. Measurement of micro/macroalbuminuria Spot untimed urine samples were collected for measurement of albumin and creatinine. Albumin was measured in milligrams per litre and creatinine in millimoles per litre. The concentration ratio of urine albumin to creatinine expressed in micrograms per milligram was used to estimate the total daily albumin excretion. Presence of micro/macroalbuminuria was defined as a urinary albumin:creatinine ratio (ACR) of ≥17 mg/g for men and ≥25 mg/g for women based on the National Kidney Foundation's Kidney Disease Outcome Quality Initiative working group definition [5]. Exposure measurement Exposure measurements were similar in both the studies. Information on participants' demographic characteristics, educational attainment, cigarette smoking, alcohol consumption and medical history were obtained using a standardized questionnaire administered by trained personnel. Age was defined as the age at the time of clinic examination. Education was categorized into: (i) primary or lower (≤6 years), (ii) secondary (7 to 10 years) and (iii) post-secondary (≥11 years, including university education). Cigarette smoking was categorized into current smokers, former or nonsmokers and alcohol consumption into drinkers (irrespective of quantity) and nondrinkers. BMI was calculated as weight in kilograms divided by the square of height in meters (kg/m2). The average of the two systolic and diastolic BP measurements was used as the systolic and diastolic BP value. Mean arterial BP was calculated as two-thirds of the diastolic plus one-third of the systolic value. Hypertension was defined as systolic BP ≥ 140 mmHg or diastolic BP ≥ 90 mmHg or self-reported physician-diagnosed hypertension. Diabetes mellitus was defined as fasting plasma glucose ≥7 mmol/L (≥126 mg/dL) or self-reported physician-diagnosed diabetes or use of glucose-lowering medication in SP2/SCCS2 and as casual plasma glucose ≥11.1 mmol/L (≥200 mg/dL) or self-reported physician-diagnosed diabetes or use of glucose-lowering medication in SiMES. High-sensitivity CRP was assayed on a Roche/Integra 400 Analyzer (Roche Diagnostics, Rotkreuz, Switzerland) by a particle-enhanced immunoturbidimetric method. The intra- and inter-assay precisions were 0.6–1.3% and 2.3–3.1%, respectively. Serum lipids, serum creatinine, plasma glucose and CRP were measured from fasting blood samples in SP2/SCCS2 and from nonfasting blood samples in SiMES. All serum biochemistry tests were carried out at the National University Hospital Reference Laboratory. Statistical analysis The two studies were combined into a single dataset, stratified by study cohort. CRP levels were analyzed as categorical and continuous variables. For categorical analysis, we grouped CRP levels into categories based on the cut points recommended by the Centers for Disease Control and Prevention and the American Heart Association [22]. For continuous analysis (per unit change), CRP was log transformed because of its skewed distribution. We compared various characteristics of the study population within ethnic groups by analysis of variance or chi-square (χ2) tests as appropriate. We then examined the association of CRP with micro/macroalbuminuria for the whole population and for the two subpopulations (SP2/SCCS2, SiMES) in two logistic regression models. In the first model, we adjusted for age (years), sex and ethnicity (Chinese, Malay and Indian). In the second multivariable model, we additionally adjusted for education (primary/below, secondary/higher), current smoking (absent, present), alcohol consumption (absent, present), diabetes status (absent, present), mean arterial BP (mm Hg), BMI (kg/m2), total cholesterol (mmol/L), HDL cholesterol (mmol/L) and study cohort (SP2/SCCS2, SiMES). Tests for trend were performed, modeling CRP categories as an ordinal variable in the corresponding multivariable logistic regression models. To examine the consistency of the association between CRP (continuous) and micro/macroalbuminuria, we performed subgroup analysis stratified by categories of age (<60, ≥60 years), sex, ethnicity (Chinese, Malay and Indians), BMI (<25, ≥25 kg/m2), diabetes (absent, present) and hypertension (absent, present). Statistical interaction between log CRP and each of the stratifying variables was examined in the corresponding logistic regression model by including cross-product interaction terms. In supplementary analyses, we repeated the multivariable analysis in Table 2 (i) using tertiles of CRP (<0.7, 0.7–2.2, ≥2.2 mg/L), (ii) after excluding participants with diabetes mellitus or hypertension, (iii) after excluding participants with macroalbuminuria (ACR ≥ 250 µg/mg for men and ≥355 µg/mg for women) and (iv) after excluding participants with CRP levels suggestive of clinical inflammation (>10 mg/L). All analyses were performed using SAS version 9.1 (SAS Institute Inc., Cary, NC, USA). Table 2 Association between C-reactive protein and micro/macroalbuminuria in SP2/SCCS2 and SiMES Variable  Number at risk (cases)  Micro/macroalbuminuria, %  Age, sex, ethnicity-adjusted OR (95% CI)  Multivariable OR (95% CI)a  CRP categories, mg/L  <1 mg/L  2109 (306)  14.5  1.00 (Referent)  1.00 (Referent)  1–3 mg/L  1762 (407)  23.1  1.67 (1.41–1.98)  1.33 (1.11–1.60)  >3 mg/L  1256 (367)  29.2  2.43 (2.03–2.90)  1.60 (1.30–1.96)  P trend      <0.0001  <0.0001  Log-transformed CRP, mg/L  5127 (1080)  21.1  1.38 (1.30–1.46)  1.20 (1.11–1.28)  Variable  Number at risk (cases)  Micro/macroalbuminuria, %  Age, sex, ethnicity-adjusted OR (95% CI)  Multivariable OR (95% CI)a  CRP categories, mg/L  <1 mg/L  2109 (306)  14.5  1.00 (Referent)  1.00 (Referent)  1–3 mg/L  1762 (407)  23.1  1.67 (1.41–1.98)  1.33 (1.11–1.60)  >3 mg/L  1256 (367)  29.2  2.43 (2.03–2.90)  1.60 (1.30–1.96)  P trend      <0.0001  <0.0001  Log-transformed CRP, mg/L  5127 (1080)  21.1  1.38 (1.30–1.46)  1.20 (1.11–1.28)  aAdjusted for age (years), gender (male, female), ethnicity (Chinese, Malay, Indians), education (primary/below, secondary/higher), current smoking (absent, present), alcohol consumption (absent, present), diabetes status (absent, present), mean arterial blood pressure (mm Hg), study type (SiMES, SP2), body mass index (kg/m2), total cholesterol (mmol/L) and high-density lipoprotein cholesterol (mmol/L). View Large Results The median CRP was 1.1 mg/L in SP2/SCCS2, 1.8 mg/L in SiMES and 1.2 mg/L in the combined population. Table 1 shows characteristics of the study population by ethnicity and study cohort. Among the three ethnic groups, Malays in both the study cohorts were more likely to be primary/below educated and smokers; had higher levels of systolic BP, diastolic BP, mean arterial BP, BMI, total cholesterol and LDL cholesterol; and less likely to consume alcohol; Chinese were more likely to be females, and Indians had higher prevalence of diabetes, higher levels of CRP and lower levels of HDL cholesterol. In general, participants from the SiMES cohort were older and had higher prevalence of all risk factors examined compared with participants from SP2/SCCS2 cohort. Table 1 Characteristics of the study participants by race/ethnicity and by study cohort   SP2/SCCS2  SiMES    Characteristic  Chinese (n = 2763)  Malay (n = 772)  Indian (n = 698)  Malay (n = 894)  P value*  Gender, male, %  44.2  49.6  49.7  50.5  0.0007  Primary/below education, %  24.3  30.7  28.2  71.9  <0.0001  Hypertension, %  38.1  46.5  41.8  67.5  <0.0001  Diabetes, %  7.2  14.6  22.1  20.4  <0.0001  Current smoking, %  10.0  17.1  10.3  22.4  <0.0001  Alcohol consumption, %  45.9  6.5  32.1  1.0  <0.0001  Age, mean (SD), years  49.2 (11.9)  49.6 (11.6)  51.6 (11.2)  56.2 (10.4)  <0.0001  Systolic BP, mean (SD), mmHg  130.3 (20.5)  137.1 (21.0)  133.2 (21.2)  147.8 (23.9)  <0.0001  Diastolic BP, mean (SD), mmHg  77.0 (11.0)  79.6 (10.7)  78.4 (10.4)  81.8 (11.4)  <0.0001  Mean arterial BP, mean (SD), mm Hg  94.6 (13.0)  98.6 (12.9)  96.5 (12.8)  103.6 (14.2)  <0.0001  Body mass index, mean (SD), kg/m2  22.9 (3.8)  26.2 (4.8)  25.8 (4.7)  26.3 (5.0)  <0.0001  Total cholesterol, mean (SD), mmol/L  5.2 (0.9)  5.5 (1.0)  5.2 (0.9)  5.6 (1.1)  <0.0001  HDL cholesterol, mean (SD), mmol/L  1.5 (0.4)  1.4 (0.3)  1.2 (0.3)  1.3 (0.3)  <0.0001  LDL cholesterol, mean (SD), mmol/L  3.1 (0.9)  3.4 (1.0)  3.3 (0.8)  3.4 (0.9)  <0.0001  C-reactive protein (mg/L)  2.0 (4.9)  3.2 (4.6)  4.4 (8.1)  3.7 (7.2)  <0.0001    SP2/SCCS2  SiMES    Characteristic  Chinese (n = 2763)  Malay (n = 772)  Indian (n = 698)  Malay (n = 894)  P value*  Gender, male, %  44.2  49.6  49.7  50.5  0.0007  Primary/below education, %  24.3  30.7  28.2  71.9  <0.0001  Hypertension, %  38.1  46.5  41.8  67.5  <0.0001  Diabetes, %  7.2  14.6  22.1  20.4  <0.0001  Current smoking, %  10.0  17.1  10.3  22.4  <0.0001  Alcohol consumption, %  45.9  6.5  32.1  1.0  <0.0001  Age, mean (SD), years  49.2 (11.9)  49.6 (11.6)  51.6 (11.2)  56.2 (10.4)  <0.0001  Systolic BP, mean (SD), mmHg  130.3 (20.5)  137.1 (21.0)  133.2 (21.2)  147.8 (23.9)  <0.0001  Diastolic BP, mean (SD), mmHg  77.0 (11.0)  79.6 (10.7)  78.4 (10.4)  81.8 (11.4)  <0.0001  Mean arterial BP, mean (SD), mm Hg  94.6 (13.0)  98.6 (12.9)  96.5 (12.8)  103.6 (14.2)  <0.0001  Body mass index, mean (SD), kg/m2  22.9 (3.8)  26.2 (4.8)  25.8 (4.7)  26.3 (5.0)  <0.0001  Total cholesterol, mean (SD), mmol/L  5.2 (0.9)  5.5 (1.0)  5.2 (0.9)  5.6 (1.1)  <0.0001  HDL cholesterol, mean (SD), mmol/L  1.5 (0.4)  1.4 (0.3)  1.2 (0.3)  1.3 (0.3)  <0.0001  LDL cholesterol, mean (SD), mmol/L  3.1 (0.9)  3.4 (1.0)  3.3 (0.8)  3.4 (0.9)  <0.0001  C-reactive protein (mg/L)  2.0 (4.9)  3.2 (4.6)  4.4 (8.1)  3.7 (7.2)  <0.0001  SD, standard deviation; HDL, high-density lipoprotein; LDL, low-density lipoprotein. *The P value represents difference in characteristics by ethnicity based on analysis of variance or chi-square test as appropriate. View Large Table 2 presents the odds ratio (OR) and 95% confidence interval (CI) of micro/macroalbuminuria in relation to CRP for the whole population. The prevalence of micro/macroalbuminuria in the whole population was 21.1%. The prevalence increased with increasing categories of CRP (P trend <0.0001). CRP was significantly associated with micro/macroalbuminuria in both the age, sex and ethnicity adjusted model and the multivariable model additionally adjusted for education, current smoking, alcohol consumption, diabetes status, mean arterial blood pressure, BMI and total and HDL cholesterol. This positive association persisted when CRP was analyzed as a continuous variable. Table 3 shows the study-specific association between CRP and micro/macroalbuminuria. The prevalence of micro/macroalbuminuria was higher in SiMES (36.5%) than in SP2/SCCS2 (17.8%). CRP was positively associated with micro/macroalbuminuria in both the models; compared with persons with CRP concentrations <1 mg/L, the multivariable OR of micro/macroalbuminuria for persons with CRP concentrations >3 mg/L was 1.6 in both SP2/SCCS2 and SiMES. In Table 4, we examined the OR of micro/macroalbuminuria associated with increasing levels of log-transformed CRP within subgroups of age, sex, ethnicity, BMI and hypertension status. In general, the positive association between CRP and micro/macroalbuminuria was consistent among these subgroups also, and no significant interactions were detected between CRP and any of the stratifying variables. Table 3 Association between C-reactive protein and micro/macroalbuminuria by study cohort   SP2/SCCS2  SiMES  Variable  Number at risk (cases)  Micro/macro albuminuria, %  Multivariable OR (95% CI)a  Number at risk (cases)  Micro/macro albuminuria, %  Multivariable OR (95% CI)a  CRP categories, mg/L  <1 mg/L  1847 (225)  12.2  1.00 (Referent)  262 (81)  30.9  1.00 (Referent)  1–3 mg/L  1436 (292)  20.3  1.35 (1.09–1.67)  326 (115)  35.3  1.33 (0.89–1.99)  >3 mg/L  950 (237)  25.0  1.60 (1.26–2.04)  306 (130)  42.5  1.65 (1.10–2.49)  P trend      0.0001      0.02  Log-transformed CRP, mg/L  4233 (754)  17.8  1.22 (1.12–1.33)  894 (326)  36.5  1.14 (0.99–1.32)    SP2/SCCS2  SiMES  Variable  Number at risk (cases)  Micro/macro albuminuria, %  Multivariable OR (95% CI)a  Number at risk (cases)  Micro/macro albuminuria, %  Multivariable OR (95% CI)a  CRP categories, mg/L  <1 mg/L  1847 (225)  12.2  1.00 (Referent)  262 (81)  30.9  1.00 (Referent)  1–3 mg/L  1436 (292)  20.3  1.35 (1.09–1.67)  326 (115)  35.3  1.33 (0.89–1.99)  >3 mg/L  950 (237)  25.0  1.60 (1.26–2.04)  306 (130)  42.5  1.65 (1.10–2.49)  P trend      0.0001      0.02  Log-transformed CRP, mg/L  4233 (754)  17.8  1.22 (1.12–1.33)  894 (326)  36.5  1.14 (0.99–1.32)  aAdjusted for age (years), gender (male, female), ethnicity (Chinese, Malay, Indians), education (primary/below, secondary/higher), current smoking (absent, present), alcohol consumption (absent, present), diabetes status (absent, present), mean arterial blood pressure (mm Hg), body mass index (kg/m2), total cholesterol (mmol/L) and high-density lipoprotein cholesterol (mmol/L). View Large Table 4 Association between increasing CRP level and micro/macroalbuminuria within selected subgroups in SP2/SCCS2 and SiMES Stratified subgroups  Number at risk (cases)  Multivariable OR (95% CI) of micro/macroalbuminuria associated with log-transformed CRP, mg/La  P interactionb  Age, years  <60  4069 (652)  1.22 (1.11–1.33)  0.41  ≥60  1058 (428)  1.15 (1.02–1.30)    Gender  Male  2403 (568)  1.26 (1.14–1.40)  0.92  Female  2724 (512)  1.15 (1.04–1.27)    Ethnicity  Chinese  2763 (429)  1.23 (1.11–1.37)  0.50  Malay  1666 (503)  1.19 (1.07–1.33)    Indian  698 (148)  1.11 (0.90–1.36)    Body mass index (kg/m2)  <25  3133 (506)  1.16 (1.06–1.28)  0.79  ≥25  1994 (574)  1.28 (1.15–1.44)    Diabetes mellitus  Absent  4479 (738)  1.20 (1.10–1.30)  0.41  Present  648 (342)  1.20 (1.02–1.40)    Hypertension  Absent  2820 (279)  1.26 (1.11–1.43)  0.38  Present  2307 (801)  1.16 (1.06–1.26)    Stratified subgroups  Number at risk (cases)  Multivariable OR (95% CI) of micro/macroalbuminuria associated with log-transformed CRP, mg/La  P interactionb  Age, years  <60  4069 (652)  1.22 (1.11–1.33)  0.41  ≥60  1058 (428)  1.15 (1.02–1.30)    Gender  Male  2403 (568)  1.26 (1.14–1.40)  0.92  Female  2724 (512)  1.15 (1.04–1.27)    Ethnicity  Chinese  2763 (429)  1.23 (1.11–1.37)  0.50  Malay  1666 (503)  1.19 (1.07–1.33)    Indian  698 (148)  1.11 (0.90–1.36)    Body mass index (kg/m2)  <25  3133 (506)  1.16 (1.06–1.28)  0.79  ≥25  1994 (574)  1.28 (1.15–1.44)    Diabetes mellitus  Absent  4479 (738)  1.20 (1.10–1.30)  0.41  Present  648 (342)  1.20 (1.02–1.40)    Hypertension  Absent  2820 (279)  1.26 (1.11–1.43)  0.38  Present  2307 (801)  1.16 (1.06–1.26)    aAdjusted for age (years), gender (male, female), ethnicity (Chinese, Malay, Indian), education (primary/below, secondary/higher), current smoking (absent, present), alcohol consumption (absent, present), diabetes status (absent, present), mean arterial blood pressure (mm Hg), body mass index (kg/m2), total cholesterol (mmol/L), high-density lipoprotein cholesterol (mmol/L) and study centre (SiMES, SP2); the stratifying variable (gender, body mass index, diabetes, hypertension) was not adjusted in its corresponding models. bP value associated with the cross-product interaction term between the corresponding stratification variable and CRP variable in the multivariable model. View Large In supplementary analyses, when we repeated the main analysis in Table 2 with tertiles of CRP, the results remained largely the same; compared with tertile 1 of CRP (referent), the multivariable OR (95% CI) was 1.29 (1.05–1.59) in tertile 2 and 1.54 (1.25–1.90) in tertile 3 (P trend <0.0001). Excluding participants with diabetes mellitus or hypertension (n = 2651 included), the results were essentially similar; compared with persons with CRP concentrations <1 mg/L, the multivariable OR (95% CI) was 1.44 (1.16–1.77) in persons with CRP concentrations 1–3 mg/L and 1.60 (1.26–2.03) in persons with CRP concentrations >3 mg/L. Excluding participants with evidence of clinical inflammation (n = 4895 included) did not materially alter the results; compared with persons with CRP concentrations <1 mg/L, the multivariable OR (95% CI) was 1.34 (1.11–1.61) in persons with CRP concentrations 1–3 mg/L and 1.54 (1.24–1.92) in persons with CRP concentrations >3 mg/L. Excluding participants with history of cardiovascular disease (n = 4827 included) slightly accentuated the ORs; compared with persons with CRP concentrations <1 mg/L, the multivariable OR (95% CI) was 1.34 (1.12–1.62) in persons with CRP concentrations 1–3 mg/L and 1.74 (1.41–2.13) in persons with CRP concentrations >3 mg/L. Finally, excluding participants with macroalbuminuria (n = 5013 included) did not substantially alter the results; compared with persons with CRP concentrations <1 mg/L, the multivariable OR (95% CI) was 1.34 (1.10–1.62) in persons with CRP concentrations 1–3 mg/L and 1.57 (1.27–1.95) in persons with CRP concentrations >3 mg/L. Discussion The present study, combining data from two population-based cohorts, shows that elevated CRP levels are significantly associated with micro/macroalbuminuria. This association was independent of age, sex, ethnicity, education, current smoking, alcohol consumption, diabetes, BP, BMI and total/HDL cholesterol and was consistently present when CRP was analyzed as either a categorical or as a continuous variable. Furthermore, the association was consistent across the two study cohorts with similar effect estimates for micro/macroalbuminuria and in analysis stratified by age, sex, ethnicity, BMI, diabetes and hypertension status. The prevalence of micro/macroalbuminuria in the current study was 21.1%. The prevalence in SiMES was higher (36.5%) than in SP2/SCCS2 (17.8%). The higher prevalence in SiMES could be attributed to the older mean age (56.2 vs. 49.7 years) and the higher prevalence of diabetes (20% vs. 11%) and hypertension (68% vs. 40%) in the study population compared to SP2/SCCS2. The prevalence of micro/macroalbuminuria in the US National Health and Nutrition Examination Surveys 1999–2004 was 13% [14] and in Japan was 15.3% [23]. The median CRP in the current study (1.2 mg/L) was lower than that in the US population (2.1 mg/L) but higher than that observed in the Japanese population (0.5 mg/L in men and 0.4 mg/L in women). Our finding of an association between CRP and micro/macroalbuminuria is consistent with previous reports supporting an association between CRP and microalbuminuria [14,15]. We further showed that this association is evident in participants without hypertension or diabetes. Several studies have shown that CRP is associated with micro/macroalbuminuria in high-risk populations such as those with diabetes [10,11,24–26] or hypertension [13,27,28]. Kshirsagar et al. showed that elevated CRP levels are associated with microalbuminuria in the US general population [14]. Nakamura et al. reported that elevated CRP levels are positively associated with microalbuminuria in apparently healthy individuals in the general population in Japan [15]. Several mechanisms have been proposed to explain the association between CRP and micro/macroalbuminuria. Animal studies have shown that CRP administration in rat models resulted in endothelial dysfunction and impaired vasoreactivity by inhibiting endothelial nitric oxide synthase [29]. Decreased nitric oxide production promotes vasoconstriction, leucocyte adherence, platelet activation, impaired coagulation and vascular inflammation [30]. CRP also promotes proinflammatory cytokine production [31] leading to mesangial cell proliferation, matrix overproduction and increased vascular permeability resulting in albuminuria [32]. As micro/macroalbuminuria is an established marker for kidney damage and a strong, independent predictor of cardiovascular morbidity and mortality [8,9], our results suggest that at least part of the reported association between CRP and cardiovascular outcomes [3,4] may be mediated by its association with kidney disease as measured by micro/macroalbuminuria. The strengths of our study include its population-based samples, large sample size and availability of information on potential confounding factors. The present study had some limitations. Firstly, the cross-sectional nature of the study limits the capability to infer causal relationships. Secondly, micro/macroalbuminuria was defined based on a single spot urinary ACR measurement resulting in misclassification of the micro/macroalbuminuria status. Thirdly, although the data were collected in a similar manner in both the studies, sampling strategies, distribution of risk factors and age of the study population were not identical. Hence, combining data from the two study populations may have resulted in a selection bias. However, a consistent association between CRP and micro/macroalbuminuria was observed in both studies. In conclusion, data from two population-based multi-ethnic Asian cohorts show that elevated CRP levels are associated with micro/macroalbuminuria. If confirmed by future prospective studies, our findings may have clinical implications for including CRP in renal risk assessments. The authors thank the staff and participants in the Singapore Malay Eye Study (SiMES), Singapore Prospective Study Program (SP2) and the Singapore Cardiovascular Cohort Study (SCCS2) for their important contributions. Funding and Support: SP2/SCCS2 was supported by the Biomedical Research Council Grants No 03/1/27/18/216 and 08/1/35/19/550 and National Medical Research Council Grants No 0838/2004 and the Singapore Bio Imaging Consortium (C-011/2006). SiMES was funded by the National Medical Research Council (NMRC), 0796/2003 & the Biomedical Research Council (BMRC), 501/1/25-5, with support from the Singapore Prospective Study Program and the Singapore Tissue Network, A*STAR. Conflict of interest statement. None declared. References 1 Erlinger TP,  Tarver-Carr ME,  Powe NR, et al.  Leukocytosis, hypoalbuminemia, and the risk for chronic kidney disease in US adults,  Am J Kidney Dis ,  2003, vol.  42 (pg.  256- 263) Google Scholar CrossRef Search ADS PubMed  2 Fried L,  Solomon C,  Shlipak M, et al.  Inflammatory and prothrombotic markers and the progression of renal disease in elderly individuals,  J Am Soc Nephrol ,  2004, vol.  15 (pg.  3184- 3191) Google Scholar CrossRef Search ADS PubMed  3 Danesh J,  Wheeler JG,  Hirschfield GM, et al.  C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease,  N Engl J Med ,  2004, vol.  350 (pg.  1387- 1397) Google Scholar CrossRef Search ADS PubMed  4 Willerson JT,  Ridker PM.  Inflammation as a cardiovascular risk factor,  Circulation ,  2004, vol.  109 (pg.  II2- II10) Google Scholar CrossRef Search ADS PubMed  5 K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification,  Am J Kidney Dis ,  2002, vol.  39 (pg.  S1- S266) PubMed  6 Keane WF.  Proteinuria: its clinical importance and role in progressive renal disease,  Am J Kidney Dis ,  2000, vol.  35 (pg.  S97- S105) Google Scholar CrossRef Search ADS PubMed  7 Locatelli F,  Marcelli D,  Comelli M, et al.  Proteinuria and blood pressure as causal components of progression to end-stage renal failure. Northern Italian Cooperative Study Group,  Nephrol Dial Transplant ,  1996, vol.  11 (pg.  461- 467) Google Scholar CrossRef Search ADS PubMed  8 Hillege HL,  Janssen WM,  Bak AA, et al.  Microalbuminuria is common, also in a nondiabetic, nonhypertensive population, and an independent indicator of cardiovascular risk factors and cardiovascular morbidity,  J Intern Med ,  2001, vol.  249 (pg.  519- 526) Google Scholar CrossRef Search ADS PubMed  9 Hillege HL,  Fidler V,  Diercks GF, et al.  Urinary albumin excretion predicts cardiovascular and noncardiovascular mortality in general population,  Circulation ,  2002, vol.  106 (pg.  1777- 1782) Google Scholar CrossRef Search ADS PubMed  10 Festa A,  D'Agostino R,  Howard G, et al.  Inflammation and microalbuminuria in nondiabetic and type 2 diabetic subjects: The Insulin Resistance Atherosclerosis Study,  Kidney Int ,  2000, vol.  58 (pg.  1703- 1710) Google Scholar CrossRef Search ADS PubMed  11 Navarro JF,  Mora C,  Maca M, et al.  Inflammatory parameters are independently associated with urinary albumin in type 2 diabetes mellitus,  Am J Kidney Dis ,  2003, vol.  42 (pg.  53- 61) Google Scholar CrossRef Search ADS PubMed  12 Stuveling EM,  Bakker SJ,  Hillege HL, et al.  C-reactive protein modifies the relationship between blood pressure and microalbuminuria,  Hypertension ,  2004, vol.  43 (pg.  791- 796) Google Scholar CrossRef Search ADS PubMed  13 Tsioufis C,  Dimitriadis K,  Taxiarchou E, et al.  Diverse associations of microalbuminuria with C-reactive protein, interleukin-18 and soluble CD 40 ligand in male essential hypertensive subjects,  Am J Hypertens ,  2006, vol.  19 (pg.  462- 466) Google Scholar CrossRef Search ADS PubMed  14 Kshirsagar AV,  Bomback AS,  Bang H, et al.  Association of C-reactive protein and microalbuminuria (from the National Health and Nutrition Examination Surveys, 1999 to 2004),  Am J Cardiol ,  2008, vol.  101 (pg.  401- 406) Google Scholar CrossRef Search ADS PubMed  15 Nakamura M,  Onoda T,  Itai K, et al.  Association between serum C-reactive protein levels and microalbuminuria: a population-based cross-sectional study in northern Iwate, Japan,  Intern Med ,  2004, vol.  43 (pg.  919- 925) Google Scholar CrossRef Search ADS PubMed  16 Anand S.S,  Razak F,  Yi Q, et al.  C-reactive protein as a screening test for cardiovascular risk in a multiethnic population,  Arterioscler Thromb Vasc Biol ,  2004, vol.  24 (pg.  1509- 1515) Google Scholar CrossRef Search ADS PubMed  17 Hughes K,  Yeo PP,  Lun KC, et al.  Cardiovascular diseases in Chinese, Malays, and Indians in Singapore. II. Differences in risk factor levels,  J Epidemiol Community Health ,  1990, vol.  44 (pg.  29- 35) Google Scholar CrossRef Search ADS PubMed  18 Tan CE,  Emmanuel SC,  Tan BY, et al.  Prevalence of diabetes and ethnic differences in cardiovascular risk factors. The 1992 Singapore National Health Survey,  Diabetes Care ,  1999, vol.  22 (pg.  241- 247) Google Scholar CrossRef Search ADS PubMed  19 Hughes K,  Aw TC,  Kuperan P, et al.  Central obesity, insulin resistance, syndrome X, lipoprotein(a), and cardiovascular risk in Indians, Malays, and Chinese in Singapore,  J Epidemiol Community Health ,  1997, vol.  51 (pg.  394- 399) Google Scholar CrossRef Search ADS PubMed  20 Cutter J,  Tan BY,  Chew SK.  Levels of cardiovascular disease risk factors in Singapore following a national intervention programme,  Bull World Health Organ ,  2001, vol.  79 (pg.  908- 915) Google Scholar PubMed  21 Foong AW,  Saw SM,  Loo JL, et al.  Rationale and methodology for a population-based study of eye diseases in Malay people: The Singapore Malay Eye Study (SiMES),  Ophthalmic Epidemiol ,  2007, vol.  14 (pg.  25- 35) Google Scholar CrossRef Search ADS PubMed  22 Pearson TA,  Mensah GA,  Alexander RW, et al.  Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association,  Circulation ,  2003, vol.  107 (pg.  499- 511) Google Scholar CrossRef Search ADS PubMed  23 Maggiore U,  Cristol JP,  Canaud B, et al.  Comparison of 3 automated assays for C-reactive protein in end-stage renal disease: clinical and epidemiological implications,  J Lab Clin Med ,  2005, vol.  145 (pg.  305- 308) Google Scholar CrossRef Search ADS PubMed  24 Marcovecchio ML,  Giannini C,  Widmer B, et al.  C-reactive protein in relation to the development of microalbuminuria in type 1 diabetes: the Oxford Regional Prospective Study,  Diabetes Care ,  2008, vol.  31 (pg.  974- 976) Google Scholar CrossRef Search ADS PubMed  25 Abrahamian H,  Endler G,  Exner M, et al.  Association of low-grade inflammation with nephropathy in type 2 diabetic patients: role of elevated CRP-levels and 2 different gene-polymorphisms of proinflammatory cytokines,  Exp Clin Endocrinol Diabetes ,  2007, vol.  115 (pg.  38- 41) Google Scholar CrossRef Search ADS PubMed  26 Schram MT,  Chaturvedi N,  Schalkwijk CG, et al.  Markers of inflammation are cross-sectionally associated with microvascular complications and cardiovascular disease in type 1 diabetes—the EURODIAB Prospective Complications Study,  Diabetologia ,  2005, vol.  48 (pg.  370- 378) Google Scholar CrossRef Search ADS PubMed  27 Perticone F,  Maio R,  Tripepi G, et al.  Microalbuminuria, endothelial dysfunction and inflammation in primary hypertension,  J Nephrol ,  2007, vol.  20  Suppl 12(pg.  S56- S62) Google Scholar PubMed  28 Pedrinelli R,  Dell'Omo G,  Di Bello V, et al.  Low-grade inflammation and microalbuminuria in hypertension,  Arterioscler Thromb Vasc Biol ,  2004, vol.  24 (pg.  2414- 2419) Google Scholar CrossRef Search ADS PubMed  29 Guan H,  Wang P,  Hui R, et al.  Adeno-associated virus-mediated human C-reactive protein gene delivery causes endothelial dysfunction and hypertension in rats,  Clin Chem ,  2009, vol.  55 (pg.  274- 284) Google Scholar CrossRef Search ADS PubMed  30 Verma S,  Wang CH,  Li SH, et al.  A self-fulfilling prophecy: C-reactive protein attenuates nitric oxide production and inhibits angiogenesis,  Circulation ,  2002, vol.  106 (pg.  913- 919) Google Scholar CrossRef Search ADS PubMed  31 Verma S,  Li S.H,  Badiwala M.V, et al.  Endothelin antagonism and interleukin-6 inhibition attenuate the proatherogenic effects of C-reactive protein,  Circulation ,  2002, vol.  105 (pg.  1890- 1896) Google Scholar CrossRef Search ADS PubMed  32 Horii Y,  Iwano M,  Hirata E, et al.  Role of interleukin-6 in the progression of mesangial proliferative glomerulonephritis,  Kidney Inter Suppl ,  1993, vol.  39 (pg.  S71- S75) © The Author 2009. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org Oxford University Press
TI - C-reactive protein and microalbuminuria in a multi-ethnic Asian population
JF - Nephrology Dialysis Transplantation
DO - 10.1093/ndt/gfp591
DA - 2009-11-13
UR - https://www.deepdyve.com/lp/oxford-university-press/c-reactive-protein-and-microalbuminuria-in-a-multi-ethnic-asian-lt9ELuA8e5
SP - 1167
EP - 1172
VL - 25
IS - 4
DP - DeepDyve
ER -