Abdominal obesity, systolic blood pressure, and microalbuminuria in ...

International Journal of Obesity (2006) 30, 800–804 & 2006 Nature Publishing Group All rights reserved 0307-0565/06 $30.00 www.nature.com/ijo

ORIGINAL ARTICLE Abdominal obesity, systolic blood pressure, and microalbuminuria in normotensive and euglycemic Korean men Y Chang1, T Yoo2, S Ryu3,4, BY Huh2, BL Cho2, E Sung4, M Park2 and SH Yoo2 1

Medical Screening Center, Kangbuk Samsung Hospital, Sungkyunkwan University, School of Medicine, Republic of Korea; Department of Family Medicine, Seoul National University Hospital, Seoul National University, School of Medicine, Republic of Korea; 3Department of Occupational Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University, School of Medicine, Republic of Korea and 4Department of Family Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University, School of Medicine, Republic of Korea 2

Objectives: To evaluate the relationship between abdominal obesity and microalbuminuria (MA) in normotensive, euglycemic Korean men. Design: A cross-sectional study at a health screening center. Subjects: A total of 1321 healthy, normotensive Korean men, aged 20–78 years, with a fasting plasma glucose level o100 mg/dl. Measurements: Height, weight, and waist; systolic blood pressure (SBP); diastolic blood pressures (DBP); urinary albumin to creatinine ratio (ACR); fasting glucose, insulin, lipids, C-reactive protein (CRP), and white blood cell count. Waist circumference (WC) was used to indicate abdominal obesity and a single measurement of ACR was used to estimate MA. We also calculated body mass index (BMI) based on weight and height. Results: Mean BMI, WC, and SBP were significantly higher in subjects with MA than in those without (24.874.1 vs 23.872.7 kg/m2, 8679 vs 8378 cm, and 11575 vs 11277 mmHg, respectively). Multiple logistic regression analyses showed that only WC and SBP were independent predictors of MA. Conclusion: WC and SBP were positively associated with MA in normotensive and euglycemic Korean men. International Journal of Obesity (2006) 30, 800–804. doi:10.1038/sj.ijo.0803210; published online 10 January 2006 Keywords: waist circumference; abdominal obesity; microalbuminuria; insulin resistance; blood pressure; multiple logistic regression analyses

Introduction Microalbuminuria (MA) is associated with adverse health outcomes in diabetic and hypertensive adults.1 The prevalence of MA has been reported at 10–42% in persons with diabetes type 2,2 and at 11–40% in those with hypertension.3,4 An appreciable prevalence of MA (5–9.7%) has been found in those without diabetes, hypertension, or cardiovascular disease.5–7 Furthermore, MA has been used to predict cardiovascular morbidity and mortality even in apparently healthy people.5–8 However, at present it is

Correspondence: Professor T Yoo, Department of Family Medicine, Seoul National University Hospital, 28, Yunggun-dong, Chongno-gu, Seoul 110744, Republic of Korea (South Korea). E-mail: [email protected], [email protected] Received 22 March 2005; revised 19 October 2005; accepted 22 October 2005; published online 10 January 2006

unclear what underlies the relationship between MA and increased cardiovascular morbidity and mortality. Several studies have suggested that insulin resistance (IR) could be associated with MA.9–11 Studies of glucose-tolerant subjects suggested that IR is not always associated with MA.12–14 Some studies suggested that abdominal obesity is independently associated with MA,15–20 whereas others showed that abdominal obesity is not related to the albuminuria level.12,13 The latter studies may have been biased by factors such as a small sample size and the former studies did not exclude subjects with high-normal blood pressure or hypertension. Studies found association between elevated blood pressure and MA.21–23 Therefore, it is not clear if abdominal obesity alone or the presence of competing risk factors, such as high-normal blood pressure or IR, is associated with MA. Few reports examed the association between MA and abdominal obesity in apparently healthy subjects,

Abdominal obesity and microalbuminuria Y Chang et al

801 particularly euglycemic and normotensive subjects. Therefore, the present cross-sectional study of apparently healthy Korean men examined whether metabolic risk factors were associated with MA.

Methods Subjects The cross-sectional study included 1321 of 2673 Korean men, aged 20–78 years, who underwent medical screening at the Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine from January to December 2004. Women were excluded due to a low number of females in the database (o30% of the total subjects) and due to menstruation as a source for false-positive measurements of urinary albumin excretion. A total of 1352 men were excluded: 63 due to a history of diabetes mellitus, 123 due to a history of hypertension, 24 due to abnormal renal function (serum creatinine 41.3 mg/dl), 13 due to dipstick-positive proteinuria or albumin to creatinine ratio (ACR)X300 mg/mg, 36 due to a history of acute or chronic medical illness, 36 due to a history of cardiovascular disease, and 205 due to current uses of antihypertensives, antidiabetics, and antihyperlipidemics. The study also excluded 404 men with a systolic blood pressure (SBP)4120 mmHg or diastolic blood pressure 480 mmHg, and 484 with fasting glucose X100 mg/dl. According to the recent recommendation by the expert committee on the diagnosis and classification of diabetes mellitus, those with fasting glucose X100 mg/dl are not euglycemic.24 The Institutional Review Board at Kangbuk Samsung Hospital approved this study.

Measurements and definitions Height and weight were measured after an overnight fast. Body mass index (BMI) was calculated as the weight (kg) divided by the square of the height (m). Waist circumference (WC) was measured by two trained personnel, to the nearest 0.1 cm at the midpoint between the bottom of the rib cage and the top of the iliac crest with the subjects standing, their weight equally distributed on both feet, their arms at their sides, and head facing straight forward. Trained nurses measured the blood pressure with the subjects in a sitting position using a standard mercury sphygmomanometer. The first and fifth Korotkoff sounds were used to estimate the SBP and diastolic blood pressure (DBP). The average of two readings was used for all analyses. After a 12-h fast, serum glucose, total cholesterol, LDLcholesterol, triglyceride, and HDL-cholesterol levels were determined using an antoanalyzer (Advia 1650 Autoanalyzer, Bayer Diagnostics, Leverkusen, Germany). An immunonephelometry assay (Dade Behring, Marburg, Germany) was used to determine high sensitivity C-reactive protein (CRP) concentrations with a detection limit of 0.175 mg/l. The

maximum inter- and intra-assay coefficients of variation (CVs) for the range of concentrations were 5.7 and 4.4% for the CRP. The CRP values were highly skewed, and were normalized by a logarithmic transformation for all analyses. Serum insulin level was measured using an immunoradiometric assay (Biosource, Belgium). The maximum inter- and intra-assay CVs for the range of insulin concentrations were 12.2 and 4.5%, respectively. A spot morning urine sample per subject was collected after an overnight fast. Urinary creatinine level was determined using the Jaffe method. Urinary albumin concentration was measured using the radioimmunoassay (Immunotech, Beckman Coulter Company) with a detection limit of 0.5 mg/l. The variability coefficients were 1–6% for the intra-assay variability and o10% for the interassay variability. We considered persons reporting that they smoked to be current smokers. Abdominal obesity was defined as WC 40.9 m based on the Asia-Pacific criteria.25 The insulin sensitivity index was based on the Homeostasis Model Assessment (HOMA) calculated according to the formula; HOMA ¼ ½fasting insulin ðmIU=mlÞfasting glycemia ðmmol=lÞ=22:5 We calculated the urinary albumin (mg/ml) to creatinine (mg/ml) ratios (ACR), and defined MA as an ACR between 30 and 300 mg/mg.

Statistical analyses All statistical analyses were performed using the SPSS (version 10.0) software package. The results are shown as the mean7s.d., or the absolute number (percentages). w2 tests or Student’s t-tests were used to examine the differences between two proportions or means, respectively. Even if ACR was transformed into a logarithm value, logarithm-transformed ACR was not normally distributed. Therefore, categorical data analyses (logistic regression analyses of MA as the dichotomous variable) were used to estimate associations between all variables and MA. Following univariate analyses, multivariate logistic regression analyses, using a backward procedure based on the likelihood ratio, were performed to identify the independent risk factors for MA. The criterion for variable removal and entry was set up to 0.10 and 0.05, respectively. Hosmer and Lemeshow statistics were used to assess the fit of the logistic regression models. Both final models passed the goodness-of-fit test. The odds ratio (OR) with 95% confidence intervals (95% CI) was estimated. P-values o0.05 were considered significant.

Results Of the 1321 subjects, 32 (2.4%) had MA. The MA subjects had significantly higher BMI, WC, and SBP than the subjects International Journal of Obesity


802 without MA (Table 1). According to the Asia-Pacific criteria, eight (0.4%) subjects had the metabolic syndrome, and none of these eight subjects had MA. None of the subjects had the metabolic syndrome as defined by the National Cholesterol Education Program Adult Treatment Panel.26 Univariate logistic regression analyses showed a significant association between WC and SBP as independent variables and MA as dependent variable (Po0.05) (Table 2). They also showed a positive association between BMI and MA (OR 1.15; 95% CI, 1.02–1.31; P ¼ 0.03). However, WC and BMI were highly correlated (r ¼ 0.87, Po0.01 ). Since WC had a stronger association with ACR compared with the BMI (Spearman r ¼ 0.11, Po0.01 for WC; r ¼ 0.07, P ¼ 0.01 for

Table 1

BMI; age- and BMI-adjusted r ¼ 0.07, P ¼ 0.01 for WC), WC but not BMI was included in the multiple regression analyses (Table 2). The multiple logistic regression included smoking status, age, WC, abdominal obesity, SBP, DBP, triglycerides, HDLcholesterol, LDL-cholesterol, CRP, glucose, HOMA values, insulin as independent variables and MA as dependent variable. In multiple logistic regression analyses without abdominal obesity as independent variable, WC and SBP were significantly associated with MA. In analyses including abdominal obesity as an independent variable, abdominal obesity and SBP were significantly associated with MA.

Characteristics of the study subjects All subjects (n ¼ 1321)

Age (years) Current smokers, frequency (%) Body mass index (kg/m2) Waist circumference (cm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) White blood cell count ( 103/ml) C-reactive protein (mg/ ml) Creatinine (mg/dl) Glucose (mg/dl) Insulin (mIU/ml) HOMA HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Triglyceride (mg/dl)

45 364 23.8 83 112 74 6.19 0.10 1.1 90 8.69 1.93 57 119 137

Without MA (n ¼ 1289)

(711) (27.6) (72.7) (78) (77) (76) (71.68) (70.14) (70.1) (76) (73.15) (70.72) (711) (729) (781)

45 352 23.8 82 112 74 6.18 0.10 1.1 90 8.67 1.92 57 119 136

With MA (n ¼ 32)

(711) (27.3) (72.7) (78) (77) (76) (71.68) (70.14) (70.1) (76) (73.14) (70.72) (711) (729) (781)

45 12 24.8 86 115 74 6.46 0.10 1.1 91 9.44 2.13 57 124 148

(710) (37.5) (73.1) (79) (75) (76) (71.71) (70.11) (70.1) (75) (73.63) (70.83) (712) (727) (767)

P-value* 0.73 0.22 0.03 0.01 0.02 0.71 0.35 0.43 0.99 0.15 0.28 0.20 0.88 0.35 0.40

MA, microalbuminuria; HOMA, Homeostasis Model Assessement Index; HDL, high density lipoprotein; LDL, low density lipoprotein. The data are represented as a mean (7s.d.) or frequency (%). *w2 test for current smoker; Mann–Whitney rank sum tests for C-reactive protein, insulin, HOMA and triglyceride; Student’s t-tests for all other variables.

Table 2

Logistic regression models of the association of albuminuria with the explanatory factors (n ¼ 1321) Univariate logistic regression

Age (years) Current smoker Waist circumference (cm) Abdominal obesity WCp90 cm WC490 cm Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Triglyceride* (mg/dl) High-density lipoprotein cholesterol (mg/dl) Low-density lipoprotein cholesterol (mg/dl) C-reactive protein* (mg/ml) HOMA*

1.01 1.66 1.06 Reference 2.75 1.07 1.01 1.00 1.00 1.01 1.11 2.08

0.97–1.04 (0.73) 0.80–3.46 (0.18) 1.02–1.11 (0.01) F 1.33–5.70 1.01–1.13 0.95–1.07 1.00–1.01 0.97–1.03 0.99–1.02 0.78–1.60 0.78–5.55

(0.01) (0.02) (0.72) (0.40) (0.88) (0.35) (0.56) (0.14)

Model 1 Multivariate logistic regression

1.06

1.06

Model 2 Multivariate logistic regression

1.01–1.11 (0.02)

1.00–1.12 (0.04)

Reference 2.54 1.06

1.22–5.30 (0.01) 1.00–1.12 (0.04)

HOMA: homeostasis model assessement index. The table shows odds ratio and 95% confidence interval (P-value): The regression analyses of Model 1 included age, smoking status, waist circumference, systolic blood pressure, diastolic blood pressure, triglycerides, HDL-cholesterol, LDL-cholesterol, glucose, HOMA and insulin but not abdominal obesity as independent variables. The regression analyses of Model 2 added abdominal obesity as independent variable. *Log-transformed values were used for listed analyses.

International Journal of Obesity


803

Discussion In the present study of normotensive and euglycemic Korean men, abdominal obesity and SBP were significantly associated with MA in multiple logistic regression analyses adjusting for BMI, DBP, CRP, glucose, and IR. Previous studies have shown that obesity is associated with MA.22,27–29 However, in our study, WC and abdominal obesity had a significant association with MA whereas BMI had not. Similarly, several studies have suggested a differential effect according to body morphology.15–20 The prevalence of MA among the study subjects was lower than that reported for a study of the general population.5–7 This is expected following the selection criteria for the study. Also previous studies found an independent association between SBP and MA, both in non-diabetic and diabetic subjects.23,29,30 Complementarily, our study showed that a high SBP within the normotensive SBP range was associated with MA. Thus our study supports that high SBP below the cutoff limit for hypertension can influence the health negatively. With respect to other risk factors, no independent association between HOMA and MA was found in the study subjects, possibly due to the large s.e. and the wide CI. The relationship between ACR and IR in apparently healthy subjects is a matter of debate. Although several studies have suggested that obesity and hyperinsulinemia are associated with an increased albumin excretion rate and that IR might be related to MA among the non-diabetic population,9–11 most of these studies were performed without excluding prediabetic subjects. On the other hand, contradictory results derived from apparently healthy, glucose-tolerant subjects suggest that MA is not necessarily related to IR.12–15 In the present study, apparently healthy subjects were carefully selected based on a fasting plasma glucose o100 mg/dl and having a normal blood pressure. The data supports a lack of relationship between MA and IR as assessed from the HOMA model. Likewise, CRP and triglyceride showed no significantly independent associations between those with and without MA, possibly due to the large s.e. and wide CI. Although the selection of exclusively healthy subjects might have attenuated the relationships between all risk factors and MA, significant associations between SBP and WC on MA were still shown. Therefore, abdominal obesity and high SBP can be more important in the early phase of renal dysfunction than diastolic blood pressure, IR, glucose levels, triglycerides, and low-grade inflammation measured as CRP. The study has limitations. First, abdominal adiposity was not assessed by more accurate methods, such as computed tomography, dual-energy X-ray absorptiometry, and magnetic resonance imaging as they are costly and not practical for routine clinical practice. They have, thus yet, been used mainly for research purposes. Second, we undertook only a single blood pressure measurement that does not take into account day-to-day variability and white coat blood pressure

effect. The use of a single urine measurement might have caused misclassification.31 The definition of IR in this study was based on only a single insulin measurement. We estimated IR based on the insulin levels and HOMA analyses, not on euglycemic insulin clamp analyses.

Conclusion Our findings support that even in an Asian normotensive, euglycemic male population, WC and SBP appear to be independent predictors of cardiovascular risk markers such as MA. Therefore, clinicians should be observant regarding abdominal obesity and SBP even in men with or without hypertension and diabetes.

Acknowledgements We thank Dr Lina Kim, (Seoul, Korea) for her help with the revision.

References 1 Gerstein HC, Mann JF, Pogue J, Dinneen SF, Halle JP, Hoogwerf B et al. Prevalence and determinants of microalbuminuria in highrisk diabetic and nondiabetic patients in the Heart Outcomes Prevention Evaluation Study. The HOPE Study Investigators. Diabet Care 2000; 23 (Suppl. 2): B35–B39. 2 Deferrari G, Repetto M, Calvi C, Ciabattoni M, Rossi C, Robaudo C. Diabetic nephropathy: from micro- to macroalbuminuria. Nephrol Dial Transplant 1998; 13 (Suppl. 8): 11–15. 3 Rosa TT, Palatini P. Clinical value of microalbuminuria in hypertension. J Hypertens 2000; 18: 645–654. 4 Bigazzi R, Bianchi S, Campese VM, Baldari G. Prevalence of microalbuminuria in a large population of patients with mild to moderate essential hypertension. Nephron 1992; 61: 94–97. 5 Jones CA, Francis ME, Eberhardt MS, Chavers B, Coresh J, Engelgau M et al. Microalbuminuria in the US population: third National Health and Nutrition Examination Survey. Am J Kidney Dis 2002; 39: 445–459. 6 Romundstad S, Holmen J, Hallan H, Kvenild K, Kruger O, Midthjell K. Microalbuminuria,cardiovascular disease and risk factors in a nondiabetic/nonhypertensive population. The NordTrøndelag Health Study (HUNT, 1995–97), Norway. J Intern Med 2002; 252: 164–172. 7 Hillege HL, Fidler V, Diercks GF, van Gilst WH, de Zeeuw D, van Veldhuisen DJ, et al., for the Prevention of Renal Vascular End Stage Disease (PREVEND) Study Group. Urinary albumin excretion predicts cardiovascular noncardiovascular mortality in general population. Circulation 2002; 106: 1777–1782. 8 Yuyun MF, Khaw KT, Luben R, Welch A, Bingham S, Day NE et al. Microalbuminuria, cardiovascular risk factors and cardiovascular morbidity in a British population: the EPIC-Norfolk populationbased study. Eur J Cardiovasc Prev Rehabil 2004; 11: 207–213. 9 Forsblom CM, Forsblom CM, Eriksson JG, Ekstrand AV, Teppo AM, Taskinen MR et al. Insulin resistance and abnormal albumin excretion in non-diabetic first-degree relatives of patients with NIDDM. Diabetologia 1995; 38: 363–369. 10 Kim YI, Kim CH, Choi CS, Chung YE, Lee MS, Lee SI et al. Microalbuminuria is associated with the insulin resistance



804 11

12

13

14

15

16

17 18

19

20

syndrome independent of hypertension and type 2 diabetes in the Korean population. Diabet Res Clin Pract 2001; 52: 145–152. Mykkanen L, Zaccaro DJ, Wagenknecht LE, Robbins DC, Gabriel M, Haffner SM. Microalbuminuria is associated with insulin resistance in nondiabetic subjects: the insulin resistance atherosclerosis study. Diabetes 1998; 47: 793–800. Hoffmann IS, Jimenez E, Cubeddu LX. Urinary albumin excretion in lean, overweight and obese glucose tolerant individuals: its relationship with dyslipidaemia, hyperinsulinaemia and blood pressure. J Hum Hypertens 2001; 15: 407–412. Nielsen S, Jensen MD. Relationship between urinary albumin excretion, body composition, and hyperinsulinemia in normotensive glucose-tolerant adults. Diabet Care 1999; 22: 1728–1733. Srinivasan SR, Myers L, Berenson GS. Risk variables of insulin resistance syndrome in African-American and Caucasian young adults with microalbuminuria: the Bogalusa Heart Study. Am J Hypertens 2000; 13: 1274–1279. Pedrinelli R, Dell’Omo G, Giampietro O, Giorgi D, Di Bello V, Bandinelli S et al. Dissociation between albuminuria and insulinaemia in hypertensive and atherosclerotic men. J Hum Hypertens 1999; 13: 129–134. Cirillo M, Senigalliesi L, Laurenzi M, Alfieri R, Stamler J, Stamler R et al. Microalbuminuria in nondiabetic adults: relation of blood pressure, body mass index, plasma cholesterol levels, and smoking: The Gubbio Population Study. Arch Intern Med 1998; 158: 1933–1939. Knight EL, Kramer HM, Curhan GC. High-normal blood pressure and microalbuminuria. Am J Kidney Dis 2003; 41: 588–595. Pinto-Sietsma SJ, Navis G, Janssen WM, de Zeeuw D, Gans RO, de Jong PE. A central body fat distribution is related to renal function impairment, even in lean subjects. Am J Kidney Dis 2003; 41: 733–741. Basdevant A, Cassuto D, Gibault T, Raison J, Guy-Grand B. Microalbuminuria and body fat distribution in obese subjects. Int J Obes Relat Metab Disord 1994; 18: 806–811. Metcalf PA, Scragg RK, Dryson E. Associations between body morphology and microalbuminuria in healthy middle-aged European, Maori and Pacific Island New Zealanders. Int J Obes Relat Metab Disord 1997; 21: 203–210.


21 Mulyadi L, Stevens C, Munro S, Lingard J, Bermingham M. Body fat distribution and total body fat as risk factors for microalbuminuria in the obese. Ann Nutr Metab 2001; 45: 67–71. 22 Rosenbaum P, Gimeno SG, Sanudo A, Franco LJ, Ferreira SR, Japanese-Brazilian Diabetes Study Group. Independent impact of glycemia and blood pressure in albuminuria on high-risk subjects for metabolic syndrome. Clin Nephrol 2004; 61: 369–376. 23 Palaniappan L, Carnethon M, Fortmann SP. Association between microalbuminuria and the metabolic syndrome: NHANES III. Am J Hypertens 2003; 16: 952–958. 24 Genuth S, Alberti KG, Bennett P, Buse J, Defronzo R, Kahn R, et al., Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Follow-up report on the diagnosis of diabetes mellitus. Diabet Care 2003; 26: 3160–3167. 25 WHO Western Pacific Region, IASO and IOTF. The Asia-Pacific Perspective: Redefining Obesity and Its Treatment. Health Communications Australia Pty Limit: Sydney, Australia, 2000. 26 Expert Panel on Detection Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–2497. 27 Bosma RJ, van der Heide JJ, Oosterop EJ, de Jong PE, Navis G. Body mass index is associated with altered renal hemodynamics in non-obese healthy subjects. Kidney Int 2004; 65: 259–265. 28 de Jong PE, Verhave JC, Pinto-Sietsma SJ, Hillege HL, PREVEND study group. Obesity and target organ damage: the kidney. Int J Obes Relat Metab Disord 2002; 26: S21–S24. 29 Martinez MA, Moreno A, Aguirre de Carcer A, Cabrera R, Rocha R, Torre A, et al., MAPA – Madrid Working Group. Frequency and determinants of microalbuminuria in mild hypertension: a primary-care-based study. J Hypertens 2001; 19: 319–326. 30 Wang W, Zhao D, Liu J, Sun JY, Wu GX, Zeng ZC et al. A prospective study of relationship between blood pressure and 10-year cardiovascular risk in a Chinese cohort aged 35–64 years. Zhonghua Nei Ke Za Zhi 2004; 43: 730–734. 31 American Diabetes Association. Diabetic nephropathy. Diabet Care 2003; 23: S421–S426.