ORIGINAL ARTICLE HIP CIRCUMFERENCE IS AN IMPORTANT PREDICTOR OF PLASMA C-REACTIVE PROTEIN LEVELS IN OVERWEIGHT AND OBESE TAIWANESE WOMEN Hei-Jen Jou 1,2, I-Ping Hsu 3, Pei-Ying Ling 1, Chia-Tzu Lin 1, Suei-Tsau Tsai 4, Hui-Ting Huang 5, Shiao-Chi Wu 6 * Department of Obstetrics and Gynecology, 1 Taiwan Adventist Hospital and 2 National Taiwan University Hospital, 3 Department of Child and Family Studies, Fu-Jen Catholic University, Departments of 4 Pediatrics and 5 Internal Medicine, Taiwan Adventist Hospital, and 6 Institute of Health and Welfare Policy, National Yang-Ming University, Taipei, Taiwan. SUMMARY Objective: To investigate the relationship between serum levels of C-reactive protein (CRP) with obesity, insulin resistance and metabolic syndrome in overweight or obese Taiwanese women. Materials and Methods: Analysis was performed on data from 59 overweight or obese women enrolled in a cross-sectional study. Pearson s correlation coefficients were determined between CRP and simple anthropometric indices, fasting insulin, homeostasis model assessment (HOMA) and serum lipid levels. Forward stepwise regression procedure was used to find a best subset of predictor variables for CRP. Results: CRP was found to have significantly positive correlation with body weight, waist circumference, hip circumference, body mass index, log of fasting glucose levels, fasting insulin levels and HOMA. Forward stepwise regression procedure showed that the best predictor for CRP was hip circumference. Conclusion: The correlation between CRP and insulin resistance or abnormal lipid levels is attributed to their strong correlations with obesity. [Taiwanese J Obstet Gynecol 2006;45(3):215 220] Key Words: body mass index, C-reactive protein, hip circumference, insulin resistance, metabolic syndrome, obesity Introduction Recent studies have demonstrated a relationship between C-reactive protein (CRP), a sensitive marker of systemic inflammation and the pathogenesis of atherosclerosis [1 6]. Elevated plasma concentration of CRP is associated with an increased risk of coronary artery disease, including stroke, myocardial infarction (MI) and peripheral vascular disease [1 7]. However, the biologic mechanism underlying this relationship is unknown and remains to be determined. *Correspondence to: Dr Shiao-Chi Wu, Institute of Health and Welfare Policy, National Yang-Ming University, 155, Section 2, Li-Loung Street, Shih Pai, Taipei 112, Taiwan. E-mail: scwu@ym.edu.tw Accepted: April 14, 2006 Adipose tissue is not only a storage tissue for fat, but it also secretes a variety of products that can exert both local and systemic effects, including leptin, tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) [8]. Elevation in serum concentrations of IL-6, a powerful inducer of inflammatory reaction, causes low-grade systemic inflammatory reaction with elevated acute phase reactants, such as CRP [1]. Plasma CRP levels are not only strongly associated with body mass index (BMI) [9], but also with obesity-related diseases, including insulin resistance, diabetes mellitus, polycystic ovarian syndrome and hyperlipidemia [10 13]. The prevalence of obesity and obesity-related diseases has increased dramatically in recent years, and has become a serious public health problem. Using the definition for Taiwanese population, the prevalence of overweight (BMI 24 and BMI < 27 kg/m 2 ) and obesity (BMI 27 kg/m 2 ) in 1993 1996 were 22.9% and Taiwanese J Obstet Gynecol September 2006 Vol 45 No 3 215
H.J. Jou, et al 10.5% for men and 20.3% and 13.2% for women, respectively [14]. Several studies have suggested that a moderate increase in BMI makes Asians more prone to insulin resistance and related diseases [15,16], and CRP has a potential role in the risk assessment of obesityrelated comorbidity [5]. However, there is currently no such data available in Taiwanese women. We therefore evaluated the relationship between plasma CRP levels with obesity, insulin resistance and metabolic syndrome in Taiwanese women. Materials and Methods Study subjects A cross-sectional study was conducted between March 2003 and February 2005 in the Women s Weight Control Clinic at Taiwan Adventist Hospital. The study was approved by the institutional research board. Information was obtained on patient history and symptoms of coronary heart disease, endocrine disease and smoking. Patients suffering from any acute infection or any chronic systemic disease (such as heart disease, chronic renal disease, thyroid disease, etc.) at the time of examination, as well as current cigarette smoking, were excluded from the present study. However, women with hypertension, elevated glucose levels, impaired fasting glucose and dyslipidemia were not excluded, because these conditions were thought to be related to obesity. Collected data included age, body weight (BW), body height (BH), waist circumference, hip circumference, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), serum levels of CRP, fasting glucose levels, fasting insulin levels, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and triglycerides (TG). A homeostasis model assessment (HOMA) was used to evaluate insulin resistance as described by Matthews et al [6]. Insulin resistance was defined as HOMA 3.8. The measurements for anthropometric indices were carried out at the Physical Examination Center, Taiwan Adventist Hospital, by an internist. Overweight was defined as BMI 24 kg/m 2 and < 27 kg/m 2, and obesity as BMI 27 kg/m 2 [14]. Laboratory measurements Blood samples were obtained after an overnight fasting period of 8 hours. Levels of fasting blood glucose, TG, HDL-C and LDL-C were measured using commercial kits on an automated analyzer (Synchron CX9; Beckman Coulter Co., Fullerton, CA, USA). Insulin levels were also determined using commercial kits on an automated analyzer (Access Immunoassay System; Beckman Coulter Co.). The mean of three separate readings of insulin levels taken within 15 minutes was used for analysis. CRP concentration was measured by a solidphase, chemiluminescent immunometric assay (Immulite 2000; Diagnostic Product Corp., Los Angeles, CA, USA) with a sensitivity of 0.01 mg/dl. Log of fasting glucose levels was used for analysis due to a skewness of > 2 in the distribution of fasting glucose levels. Statistical analysis Relationships between CRP and the various parameters were investigated using Pearson s correlation analysis. In order to investigate the effects of metabolic syndrome on serum CRP, the Wilcoxon rank-sum test was used to compare the serum levels of CRP in women who had metabolic syndrome and in those without metabolic syndrome. Similar comparison was also performed to look at the impact of insulin resistance on serum CRP levels. Forward stepwise regression procedure [17] was used to find a best subset of predictor variables for CRP. CRP was designated as a dependent variable, and all the variables that were significantly correlated with CRP in Pearson s correlation analysis were used as independent variables to calculate the relative contributions of each variable. The following variables were considered for the model: BW, waist circumference, hip circumference, BMI, log of fasting glucose levels, fasting insulin levels and HOMA. A value of p < 0.05 was considered to be statistically significant. Only variables that had p < 0.05 were considered to be outcome variables in the final fitted model. Odds ratios and 95% confidence intervals of the development of metabolic syndrome and insulin resistance when a CRP level 0.6 were also calculated. Results A total of 59 cases were included in this study. The basic characteristics of the study population are listed in the Table. BMI ranged from 24.3 to 57.5 kg/m 2 ; 69% (41/59) of subjects had BMI 27. CRP varied from 0.04 to 2.18 mg/dl, with six subjects having a CRP level > 1.0 mg/dl. Waist hip ratio (WHR) ranged from 0.71 to 1.03. HOMA varied from 0.02 to 8.52; 20 women (33%) had insulin resistance with HOMA 3.8. Fasting serum insulin levels ranged from 0.01 to 32.3 IU/mL. The results of Pearson s correlation between CRP and dependent variables are also shown in the Table. CRP was significantly positively correlated with anthropometric indices, including BW (r = 0.68, p < 0.01), BMI (r = 0.67, p < 0.01), waist circumference (r = 0.63, 216 Taiwanese J Obstet Gynecol September 2006 Vol 45 No 3
Hip Circumference and Plasma CRP Levels Table. Basic characteristics of the study population and Pearson s correlation of the variables Mean (SD) Pearson s correlation CRP Age BW BH Waist Hip BMI WHR Insulin Log Glu HOMA HDL-C LDL-C SBP DBP CRP 0.51 (0.45) Age (yr) 37.4 (10.5) 0.08 BW (kg) 79.1 (17.2) 0.68* 0.31* BH (cm) 158.7 (5.6) 0.11 0.32 0.37* Waist (cm) 95.0 (14.0) 0.63* 0.20 0.88* 0.21 Hip (cm) 110.8 (11.4) 0.72* 0.31 0.94* 0.24 0.86* BMI 31.4 (6.4) 0.67* 0.21 0.93* 0.02 0.86* 0.91* WHR 0.86 (0.07) 0.19 0.05 0.37* 0.09 0.71* 0.26 0.38* Insulin (IU/mL) 11.8 (7.5) 0.35* 0.20 0.55* 0.18 0.44* 0.45* 0.53* 0.22 Log Glu (mg/dl) 101.0 (19.0) 0.38* 0.37* 0.33 0.15 0.31 0.29 0.43* 0.20 0.23 HOMA 3.0 (2.0) 0.47* 0.08 0.61* 0.08 0.50* 0.52* 0.64* 0.25 0.93* 0.54* HDL-C (mg/dl) 44.6 (8.9) 0.14 0.33 0.30 0.02 0.368 0.31 0.31 0.25 0.31 0.08 0.26 LDL-C (mg/dl) 120.6 (30.3) 0.14 0.34* 0.10 0.03 0.13 0.15 0.11 0.04 0.11 0.20 0.02 0.39* SBP (mmhg) 132.6 (18.7) 0.18 0.11 0.43* 0.01 0.49* 0.42* 0.48* 0.35* 0.24 0.30 0.32 0.24 0.04 DBP (mmhg) 80.1 (12.5) 0.10 0.18 0.30 0.13 0.29* 0.29 0.38* 0.15 0.18 0.32 0.28 0.09 0.21 0.80* TG (mg/dl) 115.7 (62.8) 0.17 0.04 0.09 0.01 0.05 0.08 0.09 0.00 0.37* 0.10 0.35* 0.12 0.16 0.20 0.21 *p < 0.01; p < 0.05. SD = standard deviation; CRP = C-reactive protein; BW = body weight; BH = body height; Waist = waist circumference; Hip = hip circumference; BMI = body mass index; WHR = waist hip ratio; Insulin = fasting insulin; Log Glu = log(fasting glucose); HOMA = homeostasis model assessment; HDL-C = high-density lipoprotein cholesterol; LDL-C = low-density lipoprotein cholesterol; SBP = systolic blood pressure; DBP = diastolic blood pressure; TG = triglyceride. Taiwanese J Obstet Gynecol September 2006 Vol 45 No 3 217
H.J. Jou, et al p < 0.01), hip circumference (r = 0.72, p < 0.01), but not WHR. Variables associated with insulin resistance, including log of fasting glucose, fasting insulin and HOMA, were also significantly correlated with CRP levels. There were no associations between CRP and blood pressure, cholesterol and TG. The relationships of CRP with hip circumference and HOMA are presented using scatter plots in the Figure. Women with BMI 27 kg/m 2 had significantly higher CRP (0.62 ± 0.23 vs. 0.28 ± 0.05 mg/dl, p < 0.001), HOMA (3.66 ± 4.35 vs. 1.56 ± 1.65, p < 0.001) and fasting insulin (14.1 ± 50.9 vs. 6.65 ± 30.9 IU/mL, p < 0.001) than women with BMI between 24 and 27 kg/m 2. Women with insulin resistance had higher CRP than women without insulin resistance (0.76 ± 0.59 vs. 0.40 ± 0.33 mg/dl). In order to find a best subset of predictor variables for CRP, all variables that were significantly correlated with CRP, including BW, waist circumference, hip circumference, BMI, log of fasting glucose, fasting insulin and HOMA, were used as independent variables for forward stepwise regression. Forward stepwise regression showed that the best predictor variable for CRP was hip circumference, with CRP = 2.66 ± 0.03 hip circumference (r 2 = 0.52, p < 0.001). Hip circumference explained 52% of the variance of CRP. Discussion In the present study, we have shown that plasma CRP, a sensitive marker of inflammation that has been previously reported to be associated with cardiovascular disease, is positively correlated with obesity, insulin resistance and metabolic syndrome in Taiwanese women. To date, most studies have been conducted in Western populations, with higher BMI and lower fat percentage than Chinese or other Asian groups [18]. This study is A 2.5 C-reactive protein (mg/dl) 2 1.5 1 0.5 0 80 90 100 110 120 130 140 150 160 170 Hip circumference (cm) B 2.5 C-reactive protein (mg/dl) 2 1.5 1 0.5 0 0 1 2 3 4 5 6 7 8 9 HOMA Figure. Relationships between C-reactive protein and (A) hip circumference; and (B) homeostasis model assessment (HOMA); presented as scatter plots. 218 Taiwanese J Obstet Gynecol September 2006 Vol 45 No 3
Hip Circumference and Plasma CRP Levels the first to report such associations in adult women in Taiwan. Previous studies have shown that elevation in plasma CRP may be associated with obesity and cardiovascular disease [1,11]. The association between CRP levels and BMI [19] has been previously demonstrated. Another interesting but less investigated issue is the relationship between CRP and fasting insulin levels. Previous reports support the association between CRP and insulin resistance [11]. It has been suggested that a chronic inflammatory process might be an important contributor to the glucose intolerance. Elevated levels of CRP and IL-6 predict the subsequent development of type II diabetes [10,12]. Festa et al showed that chronic subclinical inflammation was part of the insulin resistance syndrome and CRP was independently related to insulin sensitivity [10]. The present study has shown a strong association between CRP levels and anthropometric indices, including BW, waist circumference, hip circumference and BMI, but not WHR. The present study also showed strong associations between CRP and fasting glucose, fasting insulin and HOMA. In addition, strong correlations between BW, BMI, waist circumference, hip circumference, serum insulin and HOMA were noted. Women with insulin resistance had higher CRP levels. Furthermore, women with higher CRP levels also had a higher incidence of insulin resistance. However, the relationship between CRP and insulin is controversial. Because obesity is intimately related to an increase in fasting insulin level, it is possible that the relationship between CRP and insulin is indirect. The relationship between fasting insulin level and CRP could present an association produced by a co-related variable, such as obesity. Obesity-related insulin resistance may be, at least in part, a chronic inflammatory disease initiated in the adipose tissue. A study performed in premenopausal women showed that obesity, but not insulin resistance, is the major determinant in serum cardiovascular risk markers [10]. Therefore, forward stepwise regression was used in the present study to find the best predictor(s) for CRP. It showed that hip circumference was the best predictor for CRP levels. The corollary would be that obesity was the most important determinant in serum cardiovascular risk markers. This result is compatible with that of Manolio et al [20]. The correlation between serum CRP levels and metabolic syndrome was also demonstrated by previous studies. In a study by Lee et al [21], significant positive correlations were identified between CRP and BMI, waist circumference, TG, blood pressure and glucose, with a significant negative correlation between CRP and HDL. Rutter et al [22] demonstrated that elevated CRP levels are related with the presence of metabolic syndrome, especially in women. The present study has demonstrated the significant positive correlations between CRP and three out of the five components of metabolic syndrome, including waist circumference, DBP and fasting glucose levels. However, blood pressure and TG were not included in the best predictor(s) for CRP. This means that the correlation between CRP and metabolic syndrome may be due to the strong correlation of CRP with obesity, rather than with other components of the metabolic syndrome. It is notable that even a mild increase in serum CRP levels may be related to a marked increase in the incidence of insulin resistance and metabolic syndrome. In the study group, a CRP level 0.6 mg/dl had a 3.56 times increased risk for metabolic syndrome, and a 3.88 times increased risk for insulin resistance, which indicated a marked increase in the risk of cardiovascular disease and type II diabetes. Several potential limitations affect the interpretation of the present study results. The most important was the relatively small sample size, which made the results of statistical analysis inconsistent. Larger scale experiment is required to obtain further conclusions. Secondly, most of the subjects in our study were overweight or obese; the results might not be generalizable to women of normal weight, or to women of other races. In conclusion, this pilot cross-sectional study has demonstrated associations between CRP, a marker of systemic inflammation, and obesity, insulin resistance and metabolic syndrome in overweight or obese women in Taiwan. The correlations between CRP and metabolic syndrome and between CRP and insulin resistance are attributed to their strong correlations with obesity. A mildly subclinical increase in serum CRP is related to a marked increase in the risk of cardiovascular disease and type II diabetes in overweight or obese women. References 1. Yudkin JS, Kumari M, Humphries SE, Mohamed-Ali V. Inflammation, obesity, stress and coronary heart disease: is interleukin-6 the link? Atherosclerosis 2000;148:209 14. 2. Blake GJ, Ridker PM. Novel clinical markers of vascular wall inflammation. Circ Res 2001;89:763 71. 3. Lemieux I, Pascot A, Prud Homme D, et al. Elevated C-reactive protein: another component of the atherothrombotic profile of abdominal obesity. Arterioscler Thromb Vasc Biol 2001;21:961 7. 4. Ridker PM, Glynn RJ, Hennekens CH. C-reactive protein adds to the predictive value of total and HDL cholesterol in determining risk of first myocardial infarction. Circulation 1998;97:2007 11. Taiwanese J Obstet Gynecol September 2006 Vol 45 No 3 219
H.J. Jou, et al 5. Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation 1998;98:731 3. 6. Matthews D, Hosker J, Rudenski A, Naylor B, Treacher D, Turner R. Homeostasis model assessment: insulin resistance and B-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412 9. 7. Kuller LH, Tracy RP, Shaten J, Meilahn EN. Relation of C-reactive protein and coronary heart disease in the MRFIT nested case-control study. Multiple Risk Factor Intervention Trial. Am J Epidemiol 1996;44:537 47. 8. Hauner H. Secretory factors from human adipose tissue and their functional role. Proc Nutr Soc 2005;64:163 9. 9. Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB. Elevated C-reactive protein levels in overweight and obese adults. JAMA 1999;282:2131 5. 10. Festa A, D Agostino R, Howard G, Mykkanen L, Tracy RP, Haffner SM. Chronic subclinical inflammation as part of the insulin resistance syndrome. The Insulin Resistance Atherosclerosis Study (IRAS). Circulation 2000;102:42 7. 11. Yudkin JS, Stehouwer CDA, Emesis JJ, Coppack SW. C-reactive protein in healthy subjects: associations with obesity, insulin resistance, and endothelial dysfunction: a potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol 1999;19:972 8. 12. Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327 34. 13. Kelly CC, Lyall H, Petrie JR, Gould GW, Connell JM, Sattar N. Low grade chronic inflammation in women with polycystic ovarian syndrome. J Clin Endocrinol Metab 2001;86: 2453 5. 14. Lin YC, Yen LL, Chen SY, et al. Prevalence of overweight and obesity and its associated factors: findings from National Nutrition and Health Survey in Taiwan, 1993 1996. Prev Med 2003;37:233 41. 15. Banerji MA, Faridi N, Atluri R, Chaiken RL, Lebovitz HE. Body composition, leptin, and insulin resistance in Asian Indian Men. J Clin Endocrinol Metab 1999;84:137 44. 16. Chandalia M, Abate N, Garg A, Stray-Gundersen J, Grundy SM. Relationship between generalized and upper body obesity to insulin resistance in Asian Indian men. J Clin Endocrinol Metab 1999;84:137 44. 17. Neter J, Kutner MH, Nachtsheim CJ, Wasserman W. Applied Linear Statistical Models, 4 th edition. USA: Times Mirror Higher Education Group, 2000:348 53. 18. Deurenberg-Yap M, Schmidt G, van Staveren WA, Deurenberg P. The paradox of low body mass index and high body fat percentage among Chinese, Malays and Indians in Singapore. Int J Obes Relat Metab Disord 2000;24:1011 7. 19. Saijo Y, Kiyota N, Kawasaki Y, Miyazaki Y, Kashimura J, Fukuda M, Kishi R. Relationship between C-reactive protein and visceral adipose tissue in healthy Japanese subjects. Diabetes Obes Metab 2004;6:249 58. 20. Manolio TA, Savage PJ, Burke GL, et al. Correlates of fasting insulin levels in young adults: the CARDIA study. J Clin Epidemiol 1991;44:571 8. 21. Lee WY, Park JS, Noh SY, et al. C-reactive protein concentrations are related to insulin resistance and metabolic syndrome as defined by the ATP III report. Int J Cardiol 2004; 97:101 6. 22. Rutter MK, Meigs JB, Sullivan LM, D Agostino RB Sr, Wilson PW. C-reactive protein, the metabolic syndrome, and prediction of cardiovascular events in the Framingham Offspring Study. Circulation 2004;110:380 5. 220 Taiwanese J Obstet Gynecol September 2006 Vol 45 No 3