Prevalence of Peripheral Artery Disease and Associated Risk Factors in a Brazilian Rural Population : The Baependi Heart Study

Mailing Address: Rafael de Oliveira Alvim Rua Alaor de Queiroz Araújo, 135. Postal Code: 29050-245, Enseada do Suá, Vitória (ES) Brazil. E-mail: r.alvim@hotmail.com Prevalence of Peripheral Artery Disease and Associated Risk Factors in a Brazilian Rural Population: The Baependi Heart Study Rafael de Oliveira Alvim,1 Fernando Augusto Lavezzo Dias,2 Camila Maciel de Oliveira,2 Andréa Roseli Vançan Russo Horimoto,3 Anderson Zampier Ulbrich,2 José Eduardo Krieger,3 Alexandre da Costa Pereira3 Universidade Federal do Espírito Santo (UFES),1 Vitória, ES Brazil Universidade Federal do Paraná (UFPR),2 Curitiba, PR Brazil Instituto do Coração (InCor) Faculdade de Medicina da Universidade de São Paulo,3 São Paulo, SP Brazil


Introduction
Peripheral artery disease (PAD) is classically defined as a condition affecting noncardiac, nonintracranial arteries, majorly due to atherosclerosis, that leads to partial obstruction of the peripheral arteries, reducing perfusion to the tissues irrigated by these arteries. 1,2ven though PAD is asymptomatic in most patients, the disease may progress with clinical symptoms (like claudication) and eventually lead to tissue necrosis.The identification of PAD can help prevent further progression of the disease itself and additional complications, considering that this condition is a risk factor for all-cause mortality [3][4][5][6][7][8] and cardiovascular death, 3,4,7,9 including coronary artery disease 3,7 and stroke. 5,7PAD is also a major cause of quality of life impairment, which in turn, worsen even more with the progression of the disease. 10,11Therefore, information on the prevalence of PAD is important to allow proper planning in public health care.[14] Epidemiological studies have determined that the prevalence of PAD in the general population range from 4-10%; [15][16][17][18][19][20][21][22][23] however, there is a clear increase in prevalence with increasing age, 13,17,19,24 with rates as high as 20% over the age of 70 years. 19,20Data on the prevalence of PAD in the general Brazilian population are scarce.Only one large multicenter study assessed the prevalence of PAD and found a high rate (10.5%) in 1,159 individuals in the general population. 18Other studies have investigated the prevalence of PAD specifically in Japanese-Brazilians 25,26 and in a small sample of patients with diabetes. 27Therefore, other large studies are needed to better access the prevalence of PAD and understand the risk factors associated with this condition in the general Brazilian population.
The Baependi Heart Study is an ongoing Brazilian cohort study established in 2005 to investigate cardiovascular risk factors and heritability. 28The study has now expanded to include other investigations such as nocturnal polygraph, heart rate variability, pulsewave velocity, 24-hour ambulatory blood pressure monitoring, 24-hour electrocardiography (Holter), and assessment of vascular age by plethysmography.Using data collected from 2010 to 2013 from the abovementioned research project, the present study aimed to assess the prevalence of PAD and investigate associated risk factors in different age groups.

Study sample
The Baependi Heart Study is a genetic epidemiological study of cardiovascular disease risk factors with a longitudinal design, whose methodology has been previously described. 28For the present analysis, we carried out a cross-sectional evaluation of data collected in the second visit of the protocol (between 2010 and 2013).This study invited 2,072 individuals (of both genders and aged 18 -102 years), distributed across 109 families residing in the municipality of Baependi, a city in a rural area (752 km 2 , 18,072 inhabitants) located in Minas Gerais State, Southeast of Brazil.Of these, 1,634 participants underwent screening for PAD.The study protocol was approved by the ethics committee of the Hospital das Clínicas (SDC: 3485/10/074), University of São Paulo, Brazil.All procedures involved in this study are in accordance with the Declaration of Helsinki from 1975, updated in 2013.Informed consent was obtained from all participants included in the study.

Anthropometric evaluations
Anthropometric parameters were measured according to a standard protocol. 28Height was measured in centimeters and weight in kilograms using a calibrated digital balance.Body mass index (BMI) was calculated as body weight (kg) divided by squared height (m 2 ).Waist circumference was measured at the mean point between the lowest rib margin and the iliac crest with the subject standing and at the maximum point of normal expiration.Obesity was defined as a BMI ≥ 30 kg/m 2 .

Blood pressure measurements
Blood pressure was measured with a standard digital sphygmomanometer (OMRON, OMRON Eletrônica do Brasil Ltda., SP, Brazil) on the left arm after 5 minutes of rest, with the subject in the sitting position.Systolic (SBP) and diastolic blood pressures (DBP) were calculated from three readings (mean value of all measurements), with a minimal interval of 3 minutes. 28Hypertension was defined as a mean SBP ≥ 140 mmHg and/or DBP ≥ 90 mmHg and/or use of antihypertensive drug. 29

Biochemical measurements
Blood levels of triglycerides, total cholesterol, high-density lipoprotein cholesterol (HDL-C), lowdensity lipoprotein cholesterol (LDL-C), and fasting glucose were measured by standard techniques in 12-h fasting blood samples. 30Glycated hemoglobin (HbA1c) levels were determined by high-performance liquid chromatography (National Glycohemoglobin Standardization Program, USA).Diabetes mellitus was diagnosed in the presence of a fasting glucose ≥ 126 mg/dL, HbA1c ≥ 6.5%, or use of antidiabetic drugs.Hypercholesterolemia was defined as a total cholesterol level ≥ 240 mg/dL.

Assessment of risk factors and depression
Physical activity level was determined by the International Physical Activity Questionnaire -Short Form (IPAQ-SF).Sedentary lifestyle was identified based on a duration of physical activity lower than 10 minutes/ day on the previous week.

Original Article
Information regarding medical history (angina pectoris, myocardial infarction, stroke, kidney failure, and depression) and environmental risk factors such as smoking and alcohol use were evaluated through a questionnaire completed by each participant.The questionnaire was based on the World Health Organization's Multinational Monitoring Trends and Determinants in Cardiovascular Disease (MONICA) 31 project epidemiological instrument, and was applied and filled out by research assistants specifically trained for this task.

Screening of peripheral artery disease
The screening of PAD was performed with the anklebrachial index (ABI), which was measured by a single trained examiner using a sphygmomanometer (Heidji, Brazil) with a cuff suitable for the circumference of the limbs and a 10 MHz portable Doppler device (DV 610B, MEDMEGA, SP, Brazil).
3][14] The methodology of the test and the ABI classification were based on recommendations of the American College of Cardiology / American Heart Association (ACC/ AHA). 12,13ABI values between 0.91 and 1.39 were considered normal.Values ≤ 0.90 were considered compatible with PAD and values ≥ 1.40 were considered inconclusive for PAD and were excluded from the analysis.

Statistical analysis
Categorical variables were compared using the chi-square test and are presented as percentage, while continuous variables are presented as mean ± standard deviation.The normality of the data was confirmed with the Kolmogorov-Smirnov test.Unpaired Student t test was performed to analyze demographic, hemodynamic, and biochemical data according to PAD status.Since the cutoff value for PAD diagnosis based on ABI is well-established in the literature, we carried out univariate and multivariate logistic regression analyses to determine the association between PAD (ABI < 0.9) as the dependent variable and age, hypertension, diabetes, myocardial infarction, smoking, and sedentary lifestyle as predictor variables.Statistical analyses were carried out using SPSS (version 19) software (Chicago, IL, USA), with the level of significance set at 5%.

Results
A total of 1,634 individuals were screened for PAD.Seven individuals presented an ABI above 1.4 and were excluded from the analysis.Therefore, 1,627 volunteers were included in the study.The age of the participants ranged from 18 to 102 years (mean 44.9 ± 16.4 years).Table 1 shows the demographic, anthropometric, biochemical, and hemodynamic characteristics of the individuals with and without PAD, defined as an ABI equal to or lower than 0.9.Age, BMI, HbA1c, and SBP were higher in volunteers with PAD.The presence of PAD was also more frequent in elderly compared with younger individuals, and in blacks compared with whites.
Figure 1 presents the data related to the prevalence of PAD using the ABI in different age groups.Overall, the prevalence was very low (1.05%).Only one case of PAD was observed below the age of 30 years, and the prevalence of PAD increased after the fifth decade, peaking at the age of 70 years, when it reached 5.2%.The frequency of PAD by decade is presented in the Table 2.
Table 3 presents the data related to the lifestyle characteristics of the volunteers.The frequency and amount of smoking were higher in individuals with PAD.There was also a higher frequency of physically inactive volunteers in the PAD group.Table 4 presents a comparison of clinical characteristics in individuals with and without PAD.A higher prevalence of hypertension, diabetes, and obesity was observed in individuals with a diagnosis of PAD.The presence of hypercholesterolemia was not different between groups.Also, a prior history of myocardial infarction was more frequent in the PAD group.23,32 In Brazil, Makdisse et al. 18 reported in 2008 a high prevalence of PAD (10.5%) in a population with a mean age of 43.8 ± 14.7 years.The study involved volunteers in large cities (over 100 thousand inhabitants) from different Brazilian regions.Another study specifically assessed the Japanese-Brazilian population and determined a prevalence of PAD of 21.1%; 26 however, the mean age of the participants in this study was higher (about 56.5 years).The present study assessed volunteers from a city with an estimated population of 18,072 inhabitants at the time of the study.The prevalence found (around 1%) is considerably low.
Most studies assessing the prevalence of PAD have included samples of individuals older than 40 years (sometimes above the ages of 55 or 65 years).If we consider only the prevalence of PAD in individuals older than 40 years in the present study (1.7%), the data are lower but in agreement to those of other studies such as the ones performed in Sri Lanka by Weragoda et al. 32 in 2015 (3.6%, n = 2,912) and in the US by Razzouk et al. 22 in 2015 (the Life Line Screening ® , 4.1%, n = 3.67 million), and Eraso et al. 16 in 2015 (4.6%, n = 7,058), which identified prevalence rates below 5% in the general population.In a study published in 2016, Alzamora et al. 15 described a prevalence of new PDA of 4.3% (n = 2,256) in a study with a follow-up of 5 years conducted in a Spanish population aged more than 55 years and described as having low-risk factors for cardiovascular disease.
In the present study, the prevalence of PAD above the age of 50 years was 2.6%.As expected, age was independently associated with a diagnosis of PAD, as identified by the present and previous studies.Therefore, the low prevalence of PAD in the present study is partially due to the low mean age of the participants, the inclusion of volunteers with age starting at 18 years, and the fact that most volunteers (62.1%) were below the age of 50 years.,32 Volunteers with PAD in the present study were older and had higher BMI, HbA1c, and SBP levels compared  with volunteers with normal ABI.Furthermore, a higher prevalence of physical inactivity, obesity and smoking frequency and intensity (cigarettes/ day) was found in volunteers with PAD.23,32 The present study also identified a higher frequency of PAD in blacks compared with whites.This finding is aligned with recent data published by Eraso et al., 16 who assessed data from the National Health and Nutrition Examination Survey (NHANES), including 7,058 subjects above the age of 40 years.Another potential explanation for the low prevalence of PAD in the assessed population is the fact that the study was conducted in a small town where overall habits (e.g., nutritional) and exposure to risk factor (e.g., mental stress), other than the investigated here, are different from those observed in large cities.Comparing our population with the one included in the study by Makdisse et al. 18 (conducted in large Brazilian cities), our results showed a lower percentage of smokers (34.3% versus 46.7%) and physically inactive individuals (41.18% versus 64.8%).Also, the study by Makdisse et al. 18 had a larger number of individuals with chronic kidney disease (2.46% versus 6.12%), which is a well-known risk factor for PAD. 16,33on applying a multivariable logistic regression model, we determined the independent predictors associated with PAD.Only age, diabetes mellitus, smoking, and sedentary lifestyle emerged as significant   factor.Other studies have also found a weak association between PAD and hypercholesterolemia when comparing risk factors 16 or no association whatsoever. 35r study has some limitations.First, as a crosssectional analysis, a causal relationship between several cardiovascular risk factors and PAD could not be established.Second, the diagnosis of PAD in the present study was established only by ABI.Despite being a simple and inexpensive method, studies have shown that ABI has a high sensitivity (90 -97%) and specificity (98 -100%) for detection of arterial stenosis greater than 50%. 36

Conclusion
In summary, in a sample from the Brazilian population, aged 18 years and above and residing in a small rural town, PAD had a low prevalence, clearly increased with age, and was more frequent in blacks than whites.Additionally, risk factors for PAD in the investigated population were smoking, sedentary lifestyle, diabetes mellitus, and age, risk factors in the study sample.This finding is aligned with that of several other studies. 16,17,19,32,34Even though PAD is majorly caused by atherosclerotic disease, hypercholesterolemia was not found as a significant risk

Figure 1 -
Figure 1 -Prevalence of peripheral artery disease (PAD) in the overall population and stratified by age group.

Table 5
presents univariate and multivariate logistic regression models for PAD.In multivariate analysis, age, diabetes, smoking, and physical inactivity were significantly and independently associated with PAD.In terms of the number of individuals included, the present study is the largest investigation of the prevalence of PAD in a Brazilian population.The Baependi Heart Study is a Brazilian cohort study investigating cardiovascular risk factors and heritability in residents of

Table 1 -Demographic, anthropometric, biochemical, and hemodynamic characteristics of individuals with and without peripheral artery disease
Categorical variables were compared using the chi-square test, and continuous variables were compared using Student t test.PAD: peripheral artery disease; n: sample size; ABI: ankle-brachial index; BMI: body mass index; WC: waist circumference; TC: total cholesterol; HDL-C: high-density lipoprotein; LDL-C: low-density lipoprotein; HbA1c: glycated hemoglobin; SBP: systolic blood pressure; DBP: diastolic blood pressure.risk factors for PAD in the study sample were age (odds ratio [OR] = 1.08), smoking (OR = 4.01), sedentary lifestyle (OR = 3.75), and diabetes mellitus (OR = 3.07).