Association between peripheral arterial disease and creactive protein in the japanese-brazilian population

OBJECTIVE
To evaluate the relationship between peripheral arterial disease and elevated levels of C-reactive protein in the Japanese-Brazilian population of high cardiovascular risk.


METHODS
We conducted a cross-sectional study derived from a population-based study on the prevalence of diabetes and associated diseases in the Japanese-Brazilian population. One thousand, three hundred and thirty individuals aged e" 30 underwent clinical and laboratory examination, including measurement of ultrasensitive C-reactive protein. The diagnosis of peripheral arterial disease was performed by calculating the ankle-brachial index. We considered with peripheral arterial disease patients who had ankle-brachial index d" 0.9. After applying the exclusion criteria, 1,038 subjects completed the study.


RESULTS
The mean age of the population was 56.8 years; 46% were male. The prevalence of peripheral arterial disease was 21%, with no difference between genders. Data analysis showed no association between peripheral arterial disease and ultrasensitive C-reactive protein. Patients with ankle-brachial index d" 0.70 showed higher values of ultrasensitive C-reactive protein and worse cardiometabolic profile. We found a positive independent association of peripheral arterial disease with hypertension and smoking.


CONCLUSION
The association between low levels of ankle-brachial index and elevated levels of ultrasensitive C-reactive protein may suggest a relationship of gravity, aiding in the mapping of high-risk patients.


INTRODUCTION INTRODUCTION INTRODUCTION INTRODUCTION INTRODUCTION
D espite changes in lifestyle and the growing therapeutic arsenal, cardiovascular diseases remain the leading cause of morbidity and mortality, especially in emerging developed countries.Twenty-year data from the Ministry of Health on the Health System showed that cardiovascular diseases are the leading cause of death in Brazil for both genders.In 2006, they accounted for 29.4% of deaths in the country, while neoplasia caused 15.1% 1 .
Cardiovascular diseases arise by large from the atherosclerotic process, so the study and understanding of new risk factors for atherosclerosis are needed to identify individuals at risk and also for the development of new therapeutic strategies.Peripheral arterial atherosclerotic disease is increasingly prevalent in modern society due in part to the increase in life expectancy, affecting 202 million people worldwide in 2010.During the last decade there was a 28.7% increase in the prevalence of the disease in countries of low and middle income per capita, and 13.1% in high-income ones 2 .
The pathophysiology of atherosclerosis is complex and multifactorial.The initial change in this process is endothelial dysfunction resulting from various factors, such as hypertension, diabetes mellitus, hypercholesterolemia, and smoking.During all phases of the atherosclerotic process there is release of cytokines, among them interleukin-6 (IL6), which is primarily responsible for the hepatic stimulation for the release of C-reactive protein (CRP), the main acute phase protein in the inflammation process 3,4 .
Thus, atherosclerosis is regarded as a dynamic and progressive disease, arising from the combination of endothelial dysfunction and inflammation.Endothelial dysfunction is characterized by decreased production of nitric oxide and increased expression of adhesion molecules.These molecules promote the internalization of monocytes, which transform into macrophages, releasing IL6, stimulating the release of CRP, which in turn promotes the reduction Garofolo Garofolo Garofolo  Garofolo Garofolo  Association between peripheral arterial disease and creactive protein in the japanese-brazilian population 169 Rev. Col. Bras.Cir.2014; 41(3): 168-175 of nitric oxide, stimulates the release of IL-6 and the expression of endothelial adhesion molecules, perpetuating the inflammatory response 3. CRP is currently considered the main clinical marker of inflammation.Its involvement in the atherosclerotic process has been discussed for decades.
The relationship between CRP and cardiovascular disease has been investigated mainly in the coronary territory 5,6 .Observational studies have consistently shown that elevated CRP levels are associated with increased risk for coronary heart disease and its dosage aids in the prediction of disease 5,6 .However, these studies provide only limited and indirect information about the performance of the marker in predicting disease 7 .
Currently, there is great interest about the real role of CRP in atherosclerosis: a causal factor or just a mediator of disease?The association between CRP and peripheral arterial disease (PAD) has been studied, however so much less extensively than in the coronary field 8 .
Our group conducted a population-based study on the prevalence of diabetes mellitus and associated diseases in Japanese immigrants and their descendants in Brazil.Japanese populations of migrant origin have high rates of diabetes and other cardiovascular risk factors 9 .A high prevalence of chronic diseases suggests that exposure to a different lifestyle exacerbates a genetic tendency to accumulate fat, increasing cardiovascular risk.These are genetically homogeneous populations, with unfavorable cardiometabolic profile, which contributes to the study of PAD and other complications of atherosclerosis 10,11 .Therefore, this population represents an opportunity to investigate the relationship of PAD with new risk factors, such as CRP.

METHODS METHODS METHODS METHODS
The study was conducted in the Japanese-Brazilian population living in Bauru, São Paulo, Brazil.Male and female individuals, aged e" 30 years were invited to participate.Details on the recruitment and selection of the sample population were previously described by the authors 10,11 .
The study was approved by the Ethics in Research Committee of the Federal University of São Paulo, under number 1544/10, and a free and informed consent was obtained from all participants.One thousand, three hundred and thirty participants were interviewed about clinical and nutritional status and scheduled for clinical and laboratory evaluation.Data on smoking and history of previous diseases were obtained.The clinical examination included anthropometric measures, blood pressure and ankle-brachial index (ABI).Blood sample was obtained after 12 hours of fasting for laboratory testing, including ultrasensitive Creactive protein (CRP).After applying the exclusion criteria (incomplete data, CRP > 10mg/L and ABI > 1.40), 1,038 subjects completed the study.
Factors of conventional and non-conventional risk: since the population is of Asian origin, we chose to use the values of Body mass index (BMI) recommended by the Japan Society for the Study of Obesity (JASO) 12 .Values of waist-hip ratio used were those recommended by the World Health Organization (WHO) 13 .
Hypertension was diagnosed in those who, during the data collection, mentioned they have the disease and were undergoing treatment, and in those who had, during the physical examination, pressure values higher than 140x90mmHg 14 .For the diagnosis of dyslipidemia, we used the reference values recommended by the National Cholesterol Education Program (NCEP) 15 .We considered diabetic patients who reported and were receiving drug treatment for the disease, and those who were diagnosed during the study according to the criteria of the American Diabetes Association (ADA) 16 .
Concentrations of uric acid up to 6mg/dL for women and up to 7mg/dL for men were considered normal.The cutoff value for homocysteine was 15 mmol/L 17 , while for CRP, the one corresponding to the population median value, which was 1.1 mg/L.Individuals with values of CRP higher than 10mg/L were excluded.Glucose and plasma lipoproteins were determined by enzymatic methods.Concentrations of CRP were determined by chemiluminescence.
The diagnosis of PAD was performed by using an 8MHz continuous wave doppler device (Imbracios®).The ABI was calculated as the quotient of auscultated pressure in the arteries of the ankle by the highest pressure obtained in the brachial arteries.As recommended by the TASC II (Transatlantic Society Consensus) 18 , we considered abnormal a value d" 0.9 and> 1.40.We also stratified ABI into three categories: d" 0.70, from 0.71 to 0.90 and> 0.90 19 .
For the descriptive analysis, we used percentages, mean and standard deviations of the subjects' variables grouped according to the presence of PAD or to the ABI values (d" 0.70; 0.71 to 0.90; e" 0.90).We verified the existence of associations between variables through the use of chi-square and prevalence ratios (PR).For the comparison of variables mean values, according to the presence of PAD to the ABI values, we used the Student t test or analysis of variance, respectively.We applied the model of Poisson regression with robust variance in obtaining the PR of PAD according to the CRP median.We adopted a similar procedure to obtain the values of odds ratios according to the ABI values (d" 0.70; from 0.71 to 0.90; e" 0.90) and median CRP.

RESULTS RESULTS
The mean age of the 1,038 Japanese-Brazilians evaluated was 56.8 years; 46% were male.The mean values of BMI, waist-hip ratio, blood pressure, fasting Garofolo  Garofolo Garofolo  Garofolo Garofolo  Association between peripheral arterial disease and creactive protein in the japanese-brazilian population glucose, triglycerides, uric acid and homocysteine levels were significantly higher in men.Total cholesterol, LDLcholesterol, HDL-cholesterol and CRP were significantly higher in women.
Table 1 shows the prevalence of conventional risk factors, CRP and homocysteine, in the population studied.Especially among men, we observed a high frequency of smoking (p < 0.01).Arterial hypertension did not differ between genders.As for disorders of glucose tolerance, the prevalence of DM was significantly higher among men (38.7% versus 31.7%,p = 0.008), who also more often presented hypertriglyceridemia, low HDLcholesterol, hyperuricemia and hyperhomocysteinemia. Higher values of LDL-cholesterol and CRP were more frequent in women.
The prevalence of PAD was 21.1% (n = 219), with no difference between genders (19.2% vs. 22.7%), and was higher in subjects aged e" 60 years with hypertension and hyperhomocysteinemia. CRP was not associated with PAD (Table 2).In the adjusted analysis, we found a correlation between PAD and hypertension and current smoking, and no relationship of PAD with CRP (Table 3).In ABI stratified analysis, we observed a relation between PAD and hypertension and current smoking, and no relationship of PAD with CRP (Table 4).
The analysis of the means, according to the stratification of the ABI, showed higher mean CRP in patients with ABI d" 0.70, but without statistical significance.Still, the mean age of participants with ABI d" 0.70 was higher than in other categories.The same occurred with the mean fasting and two-hour glucose, homocysteine, systolic blood pressure and number of cigarettes smoked per day (Table 5).
Furthermore, 70% of individuals with ABI d" 0.70 had diabetes mellitus and increased WHR values (Table 6).

DISCUSSION
We found a high prevalence of PAD (21.1%) in this population of Japanese-Brazilians.Given that the population presented unfavorable cardiometabolic profile, this high prevalence was expected.Another study, analyzing patients at high cardiovascular risk, found high rates of PAD 20 .
The risk factors independently associated with PAD were smoking and hypertension in both genders.CRP was not independently associated with PAD.However, with the stratification of the ABI, we found higher values of this variable in those with ABI d" 0.70, but without statistical significance.Nonetheless, this subgroup of patients with ABI d" 0.70 was composed of only 20 individuals, which makes the statistical analysis difficult.Still, individuals with more advanced disease    Table 3 Table 3 Table 3 Table 3 Table 3 -Values of prevalence ratios (PR) with their respective 95% confidence intervals (95% CI) for Japanese-Brazilians according to the presence of PAD and other variables (final and initial model).

Variable
between peripheral arterial disease and creactive protein in the japanese-brazilian population (ABI d" 0.70) were those with the worst cardiometabolic profile, which can also interfere with CRP levels, a phenomenon called reverse causality, ie, the atherosclerotic plaque being a focus of inflammation, observed changes in CRP values could reflect only the inflammatory activity in the plate caused by other risk factors 21 .Elevated CRP levels may be secondary to several Table 5 Table 5 Table 5 Table 5 Table 5 -Mean values and standard deviation (SD) of demographic, anthropometric, clinical and biochemical variables of Japanese-Brazilians according to the values of the ankle-brachial index.
Nevertheless, Wensley et al., in a metaanalysis on levels of CRP and risk of coronary and cerebral disease that included 194,418 individuals, concluded that the causal relationship between atherosclerosis and CRP levels are unlikely.Although there is a linear correlation between levels of CRP and atherosclerotic disease, the relationship weakens sharply after adjusting for other cardiovascular risk factors.According to these authors, CRP may be related to the severity of atherosclerotic disease 24 .
Genetic epidemiology has assisted in obtaining evidence about the involvement of CRP in atherosclerotic disease.The concentration of CRP is a hereditary trait, and there has already been identified the presence of polymorphisms in the CRP gene capable of influencing the circulating level of this protein.These findings open a new research opportunity, since the random allocation of alleles at conception allows a balanced distribution of confounding factors between genotypes.Still, the genotype is not influenced by the presence of disease, the genetic associations then being protected from reverse causality 25 . 1 Four individuals without information 2 WHR > 0.80 for women and > 0.90 for men 3 Three individuals excluded without information 4 One subject without information 5 Homocysteine available for 819 individuals Garofolo Garofolo Association between peripheral arterial disease and creactive protein in the japanese-brazilian population With this objective, Zacho et al. studied the evolution of 51,286 individuals with CRP genotypes responsible for long-term maintenance of high plasma CRP levels and the occurrence of coronary or cerebral disease.They concluded that the polymorphism in the CRP gene was associated with high CRP plasma levels, but not with increased risk of coronary or cerebral ischemia 26 .Other studies evaluating the genetic epidemiology are consistent with the absence of a causal relationship between polymorphisms of the CRP gene with coronary disease [26][27][28] .
In summary, the published genetic studies [24][25][26][27][28] were not able to assign a causal role of CRP in atherosclerosis.The development of specific inhibitors for CRP should be of great help to arrive at this answer.Under development lies the 1,6 hexane-diphosphocholine, which has a very short half-life 29 .So far, the actual role of CRP in atherosclerosis is controversial.According to Pepys, CRP is present in atherosclerotic plaques, often collocated with the elements of the complement.The presence at the crime scene, however, is not necessarily evidence of guilt 30 .
Our findings suggest that CRP is a marker of severity or intensity of the disease.Since the CRP is released in response to an inflammatory process in the plate, the higher its value, the more intense the local inflammatory reaction.As the intensity of the inflammatory response corresponds to the intensity of atherosclerotic disease, the measurement of CRP may assist in the mapping of patients, identifying those that can evolve with worse prognosis.

Table 1 Table 1 -
Prevalence of the main cardiovascular risk factors in the Japanese-Brazilian population.

Table 2 Table 2 Table 2 Table 2 Table 2 -
Number, percentage and prevalence ratios (PR) with respective 95% confidence intervals (95% CI) for Japanese-Brazilians according to the presence of PAD and demographic, anthropometric, clinical and biochemical variables.

Table 4 Table 4 Table 4 Table 4 Table 4 -
Values of the odds ratios and their 95% confidence intervals (95% CI) for Japanese-Brazilians according to the ankle-brachial index and other variables (final and initial model).

Table 6 Table 6 Table 6 Table 6 Table 6 -
Number and percentage of Japanese-Brazilians according to the values of the ankle-brachial index and demographic, anthropometric, clinical and biochemical variables.