Anthropometric measurements and markers of renal function in adults and older adults

1 Universidade Federal de Goiás, Faculdade de Nutrição, Programa de Pós-Graduação em Nutrição e Saúde. R. 227, Qd. 68, s/n., Setor Leste Universitário, 74605-080, Goiânia, GO, Brasil. Correspondência para/Correspondence to: MRG PEIXOTO. E-mail: . 2 Universidade Federal de Goiás, Faculdade de Medicina, Departamento de Pediatria. Goiânia, GO, Brasil. 3 Universidade Federal de Goiás, Faculdade de Medicina, Departamento de Nefrologia. Goiânia, GO, Brasil. ORIGINAL | ORIGINAL


I N T R O D U C T I O N
Chronic Kidney Disease (CKD) consists of renal lesion and progressive and irreversible loss of the endocrine, glomerular, and tubular kidney functions present for three months or more with various health implications 1 .Important clinical implications appear 2 when the glomerular filtration rate is below 60 mL/minutes/1.73m².
In Brazil chronic kidney disease has an incidence of approximately 115 cases for every one million inhabitants, while its prevalence is roughly 500 cases in every one million inhabitants 3 .The increasing incidence and prevalence of CKD is directly associated with an increase in the main risk factors associated with its development and progression, such as excess weight, high blood pressure, diabetes, personal and family history of CKD, smoking, age higher than 60 years, low socioeconomic class, alcoholism, and physical inactivity [4][5][6] .
Nutritional status assessment is important for the identification of nutritional status-related problems and/or inadequacies in any life stage since it directly or indirectly influences one's health prognosis 7 .Many studies have found that excess weight and abdominal fat are risk factors for http://dx.doi.org/10.1590/1678-98652016000200005CKD [8][9][10][11] .Abdominal obesity, one of the components of the metabolic syndrome, dramatically increases the rates of metabolic, cardiovascular, and renal diseases globally 12 .
Excess weight, characterized by an increase in Body Mass Index (BMI) or waist circumference, increases the number of adipocytes in the body, which increase oxidative stress and inflammatory status, which in turn many result in renal failure associated with overweight and obesity 13,14 .
In addition to assessing nutritional status, one should also assess the main biochemical markers used for screening kidney disease.These markers include albuminuria and serum creatinine, used for estimating the glomerular filtration rate 2,15 .
Since the initial stages of CKD are asymptomatic and CKD development and progression have numerous risk factors, epidemiological, clinical, and nutritional surveillance should be conducted by primary healthcare, whose critical role is to try to contain its growing incidence and mortality 2,3 .National data on the association between anthropometric indicators and markers of renal function, and on the prevalence of CKD in the population covered by the Family Health Strategy are scarce.Thus, this study aimed to assess the association between anthropometric indicators of nutritional status and markers of renal function in adults and older adults.

M E T H O D S
This is a cross-sectional study originating from a parent project called "Mapping of chronic kidney disease and its risk factors in families covered by the Family Health Strategy of eastern Goiânia".The study was approved by the Human and Animal Medical Research Ethics Committee of the Clinics Hospital of the Universidade Federal de Goiás under Protocol nº CEPMHA/HC/UFG nº 170/09.The sample of the parent study represented the users of eight Family Health Strategy units in eastern Goiânia, who were interviewed at home.The households were selected randomly in the catchment area of each unit.The study included males and females living in the selected households aged ≥6 years covered by the Family Health Strategy.Individuals who could not walk or were hospitalized at the time of data collection were excluded.
The study sample consisted of 279 adults and older adults submitted to biochemical tests and anthropometric assessment.A posteriori calculation showed that the sample size (n=279) allowed estimating the prevalence of CKD in the study population with an error margin of 3.5%, considering an expected prevalence of 9.6% and confidence level of 95.0%.For the measures of association, the study 16 had a power of 90.0% (β=10.0%)and a confidence level of 95.0% (α=5.0%) for detecting significant correlations between the anthropometric indicators and outcomes ≥2.0.The study lifestyle variables were smoking status and regular physical activity.For the variable 'regular physical activity', individuals who practiced more than 30 minutes of leisure-time physical activity at least three times a week were classified as 'yes'; the others were classified as 'no' 18 .Smoking status was classified as follows: smoker (smokes or quit smoking less than six months ago); Rev. Nutr., Campinas, 29(2):199-209, mar./abr., 2016 Revista de Nutrição http://dx.doi.org/10.1590/1678-98652016000200005ex-smoker (quit smoking six or more months ago); and nonsmoker (never smoked) 19 .
Weight, height, and waist circumference were collected using the standard procedures 20 .The individuals were weighed and measured barefoot, wearing light clothing, by the electronic scale Plenna (São Paulo, Brazil -capacity of 150 kg and accuracy of 100 g) and portable stadiometer Seca (Cotia, São Paulo, Brazil -accuracy of 0.1 cm).Waist circumference was measured by an inelastic tape measure (accuracy of 0.1 cm) at the midpoint between the anterior superior iliac crest and the last rib.BMI was calculated by dividing body weight by the square of the height.
Body mass index was classified into two categories only, non-overweight and overweight.Adults with BMI <25 kg/m² and older adults with BMI ≤27 kg/m² were considered non-overweight.Adults with BMI ≥25 kg/m² and older adults with BMI >27 kg/m² were considered overweight 21,22 .Waist circumference was classified as recommended by the World Health Organization 22 : women with Waist circumference >80 cm and men with waist circumference >94 cm are considered at higher risk of obesity-related metabolic complications.
The percentage of Body Fat (%BF) was given by the formula proposed by Deurenberg et al. 23 : %BF= (1.2 x BMI) + (0.23 x age) -(10.8 x sex) -5.4,where BMI is measured in kg/m², age in years; and female sex = 0 and male sex = 1.The cut-off point for high %BF was 32% for females and 25% for males 24 .
Blood pressure was measured by the semiautomatic devices Omron -HEM 705 CP (Omoron Health Care, São Paulo, Brazil), using the technique standardized by the VI Brazilian Guideline for Hypertension.Blood pressure was measured in both arms at the beginning of the interview and again in the arm with the higher value at the end of the interview.Both measurements were made by the same interviewer.The analyses used the mean of the baseline and final Blood pressure measurements.Individuals with casual systolic blood pressure ≥140 mmHg and/or casual diastolic blood pressure ≥90 mmHg were considered hypertensive 25 .
Individuals with a diagnosis of diabetes, using antidiabetic drugs, or whose fasting blood glucose ≥126 were considered diabetic 26 .
The markers of renal function were: serum creatinine (for calculating the Glomerular Filtration Rate [GFR] and determining hypercreatininemia), urinary creatinine, and urinary albumin (for calculating the albumin/creatinine ratio).Albumin and creatinine were quantified by the biochemical analyzer Konelab 30 (Thermo Fisher Scientific Inc., Wattham, Massachusetts, United States), albumin by the colorimetric method (bromocresol green) and creatinine by the kinetic method.
Glomerular filtration rate was calculated by the method proposed by Cockcroft & Gault 15 , which includes the variables serum creatinine, weight, age, and sex.Any individual with a GFR <60 mL/minutes/1.73m 2 regardless of cause was considered CKD positive 1 .Microalbuminuria was defined as urinary albumin/creatinine ratio between 30 and 299 mg; macroalbuminuria was defined as urinary albumin/creatinine ratio ≥300 mg 1 .Individuals with micro-and macroalbuminuria were grouped together for data analysis because the number of individuals with macroalbuminuria was low (n=2).Hypercreatininemia was defined as serum creatinine >1.3 mg/dL 27  Multiple linear regressions assessed the independent effect of anthropometric indicators (BMI and waist circumference) on GFR and albuminuria (dependent variables) adjusted for age, mean diastolic and systolic blood pressure, and diabetes Mellitus status.The significance level was set at 5% (p<0.05).The dependent variables were transformed to a logarithmic scale to achieve normal distribution.

R E S U L T S
A total of 279 individuals were assessed, 70.9% females and 22.6% older adults.The prevalence of CKD, defined as a GFR ≤60 mL/minutes, was 8.9% in the population covered by the Family Health Strategy in east Goiânia.Micro-and macroalbuminuria were present in 34.8% of the sample.
Table 1 shows the general characteristics of the participants.The median age was 45 years and did not differ between the genders.The frequency of BMI-based overweight was 57%, which also did not differ between the genders.High waist circumference was more common in The continuous variables are presented as median and interquartile range (P 25 -P 75 ); and the categorical variables as number (percentage).Chronic kidney disease defined as glomerular filtration rate <60 mL/minutes.The differences in absolute frequencies correspond to the missing values.Regarding the study chronic noncommunicable diseases, 32 (11.6%) individuals were diabetic and 85 (30.4%) were hypertensive, with no differences between the genders.Males had higher median serum creatinine (0.93 mg/dL).
Table 2 shows the clinical and lifestyle characteristics, and markers of renal function according to BMI. High waist circumference and     GFR were more common in overweight than in non-overweight males (p<0.05).Similar results were found in females.
Table 3 shows the clinical and lifestyles characteristics, and markers of renal function according to waist circumference.The markers of renal function did not differ significantly in males.However, in females the frequency of micro-and macroalbuminuria was higher in those with high waist circumference (p=0.04).
Only body mass index was associated with albuminuria.For each 1 kg/m 2 increase in BMI, log albuminuria increased 0.028±0.01mL/minutes in females (p=0.016).

D I S C U S S I O N
Overweight based on BMI and abdominal obesity based on waist circumference were positively associated with GFR.Albuminuria was associated with high waist circumference only in females.Moreover, the study sample had a high prevalence of CKD (8.9%).This prevalence is similar to those found in other Brazilian cities.In Juiz de Fora (MG) the prevalence of CKD in individuals aged more than 18 years was 9.6% 9 , and in São Paulo (SP) the prevalence in adults was 8.4% 28 .
The prevalence of CKD in the study sample was similar to the prevalences reported by foreign studies, such as two studies from Peking (China), which found prevalences of 9.4% in 2006 29 and 10.8% in 2012 30 .In Liege (Belgium) the prevalence of CKD was 9.8% 31 .However, some studies find higher prevalences of CKD, ranging from 15.3% in a community of ethnic minorities in China 32 to 21.8% in Iran 33 .Hence, the prevalence of CKD in the global population, especially in developing countries, is already a serious public health problem 12,28,33 .
The 34.7% prevalence of micro-and macroalbuminuria in the study sample is similar to the 38.8% found in Tibet 34 .Smaller prevalences have been found in other locations, varying from 4.51 to 12.5% 28,30,33,35 .This difference between the prevalences can be explained by the fact that albuminuria increases gradually according to age and many studies did not include older adults.Król et al. 36 found that the prevalence of albuminuria increased from 8.8% in younger individuals to 32.0% in older adults, pointing out the influence of age.Brown et al. 37 reported that the prevalences of albuminuria increase when family members of CKD patients are included in a study since family history has a direct influence on kidney disease.
The present study collected data from individuals belonging to the same family and many already had CKD (8.9% of the sample), justifying the high prevalence of albuminuria.Boer et al. 38 believe that the prevalence of microalbuminuria can vary greatly if two consecutive albuminuria tests are not performed.
The present study found a frequency of hypercreatininemia of 21.5%.On the other hand, two other studies conducted in Brazil, one in Salvador (BA) 27 and another in Bambuí (MG) 39 reported hypercreatininemia prevalences of 3.1 and 5.1%, respectively.These differences may be explained by the different serum creatinine cut-off points used by the three studies -the present study used a cut-off point of 1.3 mg/dL.Additionally, antihypertensive drugs and higher age increase creatinine levels in males 27,40,41 .
Excess weight is considered a risk factor for the onset of CKD, and its prevalence has increased significantly in all continents 3 .The frequency found by this study was similar to those reported by other studies [41][42][43] .
In this study excess weight based on BMI and central obesity based on waist circumference were positively associated with high GFR.Other studies found similar results 41,[43][44][45] .
Excess weight increases sodium retention in renal tubules, reducing sodium excretion in the urine.To reduce the effects of retention in the short run, the body expands the intravascular volume, increases blood pressure and plasma flow in the kidneys, and increases the GFR, resulting in glomerular hyperfiltration, which reinforces the positive association between anthropometric indicators and GFR.However, as this scenario persists in the long run, microalbuminuria would develop followed by a decrease in GFR 46 .
Higher body weight may increase angiotensinogen expression in adipocytes, with greater formation of circulating angiotensin II and consequently, more lipogenesis.This mechanism changes blood pressure homeostasis, which may also change glomerular homeostasis 47 .
Obesity also affects the structure of the renal medulla because adipose tissue on the renal capsule, which is more developed in overweight individuals, may penetrate the renal medulla and compress the glomerular and tubular filtration systems, leading to higher blood pressure to compensate nephron compression, which in turn increases GFR and consequently, causes hyperfiltration and higher tubular reabsorption of sodium 48 .
The current study used the formula for estimating the glomerular filtration rate proposed by Cockcroft & Gault 15 , but this formula is known for not taking into account the low excretion of creatinine in obese individuals, resulting in an overestimation of the GFR of individuals with BMI >25 kg/m² 49 .In Brazil Napoli-Filho et al. 41 found that renal function estimated by the formula of Cockcroft-Gault was overestimated by 8.0% in individuals with BMI between 25 and 30 kg/m² and by as much as 29.6% in individuals with grade III obesity.
Regarding the association between anthropometric indicators and albuminuria, only BMI was associated with albuminuria, and this association occurred only in females, a finding also reported by Thoenes et al. 50.http://dx.doi.org/10.1590/1678-98652016000200005Boer et al. 38 found that BMI and waist circumference were associated with a greater likelihood of albuminuria.Clinical studies suggest that adiponectin can have a key role in the development of obesity-related albuminuria 51 .Moreover, high leptin secretion in individuals with abdominal obesity is one of the mechanisms associated with the development of albuminuria.The development of CKD varies between males and females, and Franceschini et al. 14 found that central obesity in menopausal women is associated with the risk of CKD development and progression, even in women with normal BMI.This study did not investigate albuminuria.
The cross-sectional design of the present study does not allow identifying what occurs between exposure and outcome.Other limitations include the small number of older adults in the sample and the instrument used for assessing the level of physical activity: although widely used, it is limited because of its lack of quantitative accuracy.
The high prevalence of CKD found by this population-based study reinforces the need of primary healthcare to identify and control early on the risk factors for CKD.The anthropometric indicators emphasize the importance of intensifying the strategies that prevent increases in BMI and waist circumference in adults and older adults.

C O N C L U S I O N
The study sample had a high prevalence of chronic kidney disease and excess weight.The anthropometric indicators body mass index and waist circumference were positively associated with glomerular filtration rate in both genders.
Data were collected from September 2011 to March 2013 after conducting a pilot study with 10 families.Trained interviewers accompanied by community health agents used a standardized questionnaire to collect socioeconomic, demographic, lifestyle, anthropometric, and laboratory data in home interviews.Socioeconomic class was classified as recommended by the Critério Padrão de Classificação Economica Brasil -2008 17 .The socioeconomic classes were grouped because of the small sample size as follows: class A/B, class C, and class D/E; class A/B consists of individuals with the highest income and class D/E, of individuals with the lowest income.
C O N T R I B U T O R S VAG AMADOR, AV NAGHETTINI, and ERS PEREIRA participated in data collection and analysis, literature search, and article writing.ATVS FREITAS participated in literature search and article writing.MRG PEIXOTO participated in data analysis, literature search, and article writing.
. The continuous variables with normal distribution were expressed as mean ± standard deviation, and the other variables, as median and interquartile range (25 th and 75 th percentiles).The categorical variables were expressed as absolute and relative frequencies.The associations between the categorical variables were measured by the Chi-http://dx.doi.org/10.1590/1678-98652016000200005square test.The differences between the continuous variables were measured by the Student's t test or the Mann-Whitney test.Spearman's correlation coefficient determined whether BMI and waist circumference were correlated.

Table 1 .
Sociodemographic and health characteristics of users of the Family Health Strategy in Eastern Goiânia (GO), Brazil, 2013.
* p-value of the Chi-square test (categorical variables) or Mann-Whitney test (continuous variables).

Table 2 .
Clinical and lifestyle characteristics, and markers of renal function according to the body mass index of users of the Family Health Strategy, eastern Goiânia (GO), Brazil, 2013.

Table 3 .
Clinical and lifestyle characteristics, and markers of renal function according to the waist circumference of users of the Family Health Strategy, eastern Goiânia (GO), Brazil, 2013 * Female Note: * p-value in the Chi-square test (categorical variables) or Mann-Whitney test (continuous variables).WC: Waist Circumference; SBP: Systolic Blood Pressure; DBP: Diastolic Blood Pressure; Serum creat.: Serum Creatinine; GFR: Glomerular Filtration Rate; ACR: Urinary Albumin/Creatinine Ratio.

Table 4 .
Results of the multiple linear regression models that assess whether body mass index and waist circumference are associated with the markers of renal function.Goiânia (GO), Brazil, 2013.Linear regression results are expressed as β coefficient ± standard error and standardized β coefficient (standardized measure that verifies the effects of anthropometric parameters on glomerular filtration rate).Models adjusted for age, diabetes Mellitus, and systolic and diastolic blood