Medidas antropométricas e de composição corporal como preditoras de alterações da pressão arterial em adultos
Arquivos
Data
2010-02-26
Autores
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Viçosa
Resumo
O presente trabalho objetivou avaliar a capacidade de indicadores antropométricos e de composição corporal em predizer alteração nos níveis de pressão arterial em indivíduos adultos na cidade de Viçosa, MG, além de verificar a concordância entre métodos de avaliação da composição corporal. Realizou-se estudo transversal, cuja amostra foi composta por 161 voluntários de ambos os sexos, com idade entre 20 e 59 anos. A avaliação antropométrica incluiu medidas de peso, altura, perímetros da cintura (PC), quadril e coxa e diâmetro abdominal sagital (DAS). Foram calculadas as relações cintura/ quadril (RCQ), cintura/ estatura (RCE) e cintura/ coxa (RCCx) e os índices de conicidade (ICO), do diâmetro abdominal (IDA) e de massa corporal (IMC). Para verificar o percentual de gordura corporal total (%GCT) utilizou-se a bioimpedância elétrica tetrapolar (BIA). Para comparar os métodos preditivos de gordura corporal (bioimpedância elétrica bipolar (Tanita®) e as equações estimativas de gordura corporal propostas por Deuremberg et al.(1991), Gallagher et al (2000) e Lean et al.(1996)) utilizou-se a BIA como método de referência. Aferiu-se a pressão arterial (PA) utilizando-se esfigmomanômetro aneroide, considerando como PA alterada o percentil 75 como ponto de corte. As análises estatísticas foram realizadas por meio da análise de correlação intraclasse, da análise de variância com o teste post-hoc de Tukey, do teste de Kruskall-Wallis com o teste post-hoc de Dunn s, dos coeficientes de correlação de Spearman e Pearson, da construção de curvas ROC (Receiver Operating Characteristic Curve) e da análise de concordância Bland-Altman. Identificou-se 32,3% e 43,5% dos homens com PAS e PAD alteradas, respectivamente e, 37,4% e 43,4% das mulheres com PAS e PAD superior ao percentil 75, respectivamente. Verificou-se que os indivíduos com PA alterada apresentam PC e DAS significantemente maior que os com PA desejável. O PC aferido na menor cintura diferiu significantemente dos demais locais avaliados (p<0,001) no sexo feminino; já no masculino, foi menor (p<0,05) que a cintura aferida no ponto médio. O DAS não diferiu em nenhuma das medidas para ambos os sexos. Todas as medidas do PC e do DAS apresentaram reprodutibilidade bastante elevadas (0,999-1,000). As correlações apresentadas pelo PC e pelo DAS foram moderadas, exceto o PC em relação à PAS no sexo masculino, cuja correlação foi fraca (r=0,298). Verificou-se que não houve diferença significante entre as áreas abaixo da curva para o PC e para o DAS na predição de alteração da PA em indivíduos de ambos os sexos. Os nove indicadores de obesidade avaliados foram capazes de predizer alteração na PA. Em concordância com as análises de correlação, o DAS foi o indicador que apresentou melhor capacidade preditiva da PAS e da PAD na população masculina, cujos pontos de corte foram 18,5cm e 18,4cm, respectivamente. Já na feminina, também de forma coerente com as análises de correlação, os indicadores que apresentaram maior poder discriminatório da PAS e da PAD, respectivamente, foram o IDA e o %GCT, sendo os melhores pontos de corte os valores 0,36 e 29,8%. Quanto ao diagnóstico da obesidade observou-se que, segundo o IMC, 9,9% dos indivíduos apresentaram obesidade. Entretanto, 28,6% da população estudada, possuíam percentual de gordura elevado, enquadrando-os no grupo de obesos. Dentre os métodos testados na população masculina, a equação de Gallagher et al. (2000) foi a única que não apresentou diferença estatística quando comparada à BIA; já na feminina, todas as equações diferiram significantemente. Para os homens, a Tanita® e a equação de Gallagher et al.(2000) foram os métodos que apresentaram melhor concordância. Entretanto, a Tanita® e todas as equações estudadas superestimaram de forma significante o %GCT das mulheres e nenhum dos métodos apresentou EPE<3,5%. A julgar pelos resultados encontrados, inferese que, não apenas indicadores da adiposidade abdominal, mas também os marcadores da obesidade geral apresentam aplicabilidade prática na predição de níveis pressóricos aumentados. Ressalta-se, assim, a importância da utilização de indicadores de obesidade seja abdominal ou geral, em detrimento à utilização isolada do IMC por dois motivos principais primeiro, por esse indicador ter apresentado menor eficiência em predizer alteração dos níveis pressóricos quando comparado aos acima mencionados e, segundo, por ter classificado erroneamente como eutróficos indivíduos com %GCT suficientemente elevado para classificá-los como obesos. Ressalta-se que, o uso de métodos ou equações inadequadas ao tipo de população a ser avaliada pode prejudicar o rastreamento do excesso de peso e o correto diagnóstico da obesidade. Partindo do conhecimento das alterações na pressão arterial causadas pela obesidade; ciente que indivíduos com níveis pressóricos acima do considerado desejável apresentam a mesma fisiopatologia daqueles com hipertensão arterial sistêmica e que esses indivíduos possuem grande chance de desenvolver tal morbidade, julga-se necessário a utilização de indicadores e pontos de cortes acurados a fim de detectar precocemente indivíduos predispostos a desenvolver a pré-hipertensão. Assim, doenças crônicas como a hipertensão podem ser evitadas, ou ao menos, retardadas.
This study aimed to evaluate the ability of anthropometric indicators and body composition in predicting of blood pressure disturbance in adults in Viçosa, MG, and check the agreement between methods of assessing body composition. Was conducted cross-sectional study, whose sample consisted of 161 volunteers of both sexes, aged between 20 and 59 years. Anthropometric measures included weight, height, waist circumference (WC), hip and thigh and sagittal abdominal diameter (SAD). We calculated the waist-hip ratio (WHR), waist / height (WHtR) and waist / thigh (WTR) and conicity index (CI), the abdominal diameter (ADI) and body mass index (BMI). To check the total body fat percentage (%TBF) was used the tetrapolar bioelectrical impedance analysis (BIA). To compare the predictive methods of body fat (bipolar bioelectrical impedance (Tanita ®) and the equations of body fat estimates proposed by Deuremberg et al. (1991), Gallagher et al. (2000) and Lean et al. (1996)) was used BIA as the reference method. The blood pressure (BP) was measured using a sphygmomanometer, considering as BP disturbance the 75th percentile as the cut-off point. Statistical analysis was performed using the intraclass correlation analysis, analysis of variance with Tukey post-hoc test, the Kruskal-Wallis test with post hoc Dunn's, the correlation coefficients of Spearman and Pearson construction of ROC curves (Receiver Operating Characteristic Curve) and the analysis of Bland-Altman agreement. We identified 32.3% and 43.5% of men with disturbance SBP and DBP, respectively, and 37.4% and 43.4% of women with SBP and DBP above the 75th percentile, respectively. It was found that people with BP disturbance have WC and SAD significantly higher than the desirable BP. The WC measured at lower waist differed significantly from other locations evaluated (p<0.001) in females, although male was lower (p<0.05) than the waist measured at the midpoint. The SAD did not differ in any of the measures for both sexes. All measures of the WC and the SAD showed very high reproducibility (0,999-1,000). The correlations presented by the WC and the SAD were moderate, except the WC in relation to SBP in males, which correlation was weak (r=0.298). It was found that there was no significant difference between the areas under the curve for the WC and the SAD in the prediction of BP disturbance in individuals of both sexes. The nine evaluated indicators of obesity were able to predict disturbance in BP. In agreement with the correlation analysis, the SAD was the indicator that showed the greatest predictive capacity of SBP and DBP in the male population, whose cut-off points were 18.5 cm and 18.4 cm, respectively. Although in the female is also consistent with the correlation analysis, the indicators that showed the highest discriminatory power of SBP and DBP, respectively, were the ADI and %TBF, and the best cut-off values 0.36 and 29.8%. On the diagnosis of obesity was observed that, according to BMI, 9.9% of patients had obesity. However, 28.6% of the population studied, had higher percentage of fat, framing them in the obese group. Among the methods tested in the male population, the equation of Gallagher et al. (2000) was the only one that did not show statistical difference when compared to the BIA, although in the female, all equations were significantly different. For men, the Tanita® and the equation of Gallagher et al. (2000) were the methods that showed better agreement. However, the Tanita® and all the studied equations significantly overestimated the %TBF of women and none of the methods presented SEE<3.5%. Judging by the results, it is assumed that not only indicators of abdominal adiposity, but also markers of general obesity have practical applicability in the prediction of elevated blood pressure levels. It should be noted, therefore, the importance of using obesity indicators, whether abdominal or general, over the use of BMI alone for two main reasons - first, for this indicator having less efficiency in predicting disturbance in blood pressure levels when compared to the above mentioned and, second, because it erroneously classified as normal subjects with %TBF high enough to classify them as obese. It was observed that the use of inappropriate methods or equations of the type of population to be evaluated can affect the tracking of overweight and correct diagnosis of obesity. Based on the knowledge of disturbance in blood pressure caused by obesity, aware that individuals with blood pressure levels higher than considered desirable have the same pathophysiology of those with hypertension and that these individuals have a greater chance of developing such morbidity, we think it is necessary to use indicators and accurate cut-off points in order to early detect individuals predisposed to develop pre-hypertension. Thus, chronic diseases such as hypertension can be prevented, or at least delayed.
This study aimed to evaluate the ability of anthropometric indicators and body composition in predicting of blood pressure disturbance in adults in Viçosa, MG, and check the agreement between methods of assessing body composition. Was conducted cross-sectional study, whose sample consisted of 161 volunteers of both sexes, aged between 20 and 59 years. Anthropometric measures included weight, height, waist circumference (WC), hip and thigh and sagittal abdominal diameter (SAD). We calculated the waist-hip ratio (WHR), waist / height (WHtR) and waist / thigh (WTR) and conicity index (CI), the abdominal diameter (ADI) and body mass index (BMI). To check the total body fat percentage (%TBF) was used the tetrapolar bioelectrical impedance analysis (BIA). To compare the predictive methods of body fat (bipolar bioelectrical impedance (Tanita ®) and the equations of body fat estimates proposed by Deuremberg et al. (1991), Gallagher et al. (2000) and Lean et al. (1996)) was used BIA as the reference method. The blood pressure (BP) was measured using a sphygmomanometer, considering as BP disturbance the 75th percentile as the cut-off point. Statistical analysis was performed using the intraclass correlation analysis, analysis of variance with Tukey post-hoc test, the Kruskal-Wallis test with post hoc Dunn's, the correlation coefficients of Spearman and Pearson construction of ROC curves (Receiver Operating Characteristic Curve) and the analysis of Bland-Altman agreement. We identified 32.3% and 43.5% of men with disturbance SBP and DBP, respectively, and 37.4% and 43.4% of women with SBP and DBP above the 75th percentile, respectively. It was found that people with BP disturbance have WC and SAD significantly higher than the desirable BP. The WC measured at lower waist differed significantly from other locations evaluated (p<0.001) in females, although male was lower (p<0.05) than the waist measured at the midpoint. The SAD did not differ in any of the measures for both sexes. All measures of the WC and the SAD showed very high reproducibility (0,999-1,000). The correlations presented by the WC and the SAD were moderate, except the WC in relation to SBP in males, which correlation was weak (r=0.298). It was found that there was no significant difference between the areas under the curve for the WC and the SAD in the prediction of BP disturbance in individuals of both sexes. The nine evaluated indicators of obesity were able to predict disturbance in BP. In agreement with the correlation analysis, the SAD was the indicator that showed the greatest predictive capacity of SBP and DBP in the male population, whose cut-off points were 18.5 cm and 18.4 cm, respectively. Although in the female is also consistent with the correlation analysis, the indicators that showed the highest discriminatory power of SBP and DBP, respectively, were the ADI and %TBF, and the best cut-off values 0.36 and 29.8%. On the diagnosis of obesity was observed that, according to BMI, 9.9% of patients had obesity. However, 28.6% of the population studied, had higher percentage of fat, framing them in the obese group. Among the methods tested in the male population, the equation of Gallagher et al. (2000) was the only one that did not show statistical difference when compared to the BIA, although in the female, all equations were significantly different. For men, the Tanita® and the equation of Gallagher et al. (2000) were the methods that showed better agreement. However, the Tanita® and all the studied equations significantly overestimated the %TBF of women and none of the methods presented SEE<3.5%. Judging by the results, it is assumed that not only indicators of abdominal adiposity, but also markers of general obesity have practical applicability in the prediction of elevated blood pressure levels. It should be noted, therefore, the importance of using obesity indicators, whether abdominal or general, over the use of BMI alone for two main reasons - first, for this indicator having less efficiency in predicting disturbance in blood pressure levels when compared to the above mentioned and, second, because it erroneously classified as normal subjects with %TBF high enough to classify them as obese. It was observed that the use of inappropriate methods or equations of the type of population to be evaluated can affect the tracking of overweight and correct diagnosis of obesity. Based on the knowledge of disturbance in blood pressure caused by obesity, aware that individuals with blood pressure levels higher than considered desirable have the same pathophysiology of those with hypertension and that these individuals have a greater chance of developing such morbidity, we think it is necessary to use indicators and accurate cut-off points in order to early detect individuals predisposed to develop pre-hypertension. Thus, chronic diseases such as hypertension can be prevented, or at least delayed.
Descrição
Palavras-chave
Pressão arterial, Composição corporal, Indicadores antropométricos, Blood pressure, Body composition, Anthropometric indicators
Citação
OLIVEIRA, Josie de Souza. Anthropometric measurements and body composition as predictors of blood pressure disturbance in adults. 2010. 167 f. Dissertação (Mestrado em Valor nutricional de alimentos e de dietas; Nutrição nas enfermidades agudas e crônicas não transmis) - Universidade Federal de Viçosa, Viçosa, 2010.