<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.1 20151215//EN" "https://jats.nlm.nih.gov/publishing/1.1/JATS-journalpublishing1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink" specific-use="sps-1.9" dtd-version="1.1" xml:lang="pt" article-type="research-article">
  <front>
    <journal-meta>
      <journal-id journal-id-type="publisher-id">abc</journal-id>
      <journal-title-group>
        <journal-title>Arquivos Brasileiros de Cardiologia</journal-title>
        <abbrev-journal-title abbrev-type="publisher">Arq. Bras. Cardiol.</abbrev-journal-title>
      </journal-title-group>
      <issn pub-type="ppub">0066-782X</issn>
      <issn pub-type="epub">1678-4170</issn>
      <publisher>
        <publisher-name>Sociedade Brasileira de Cardiologia - SBC</publisher-name>
        <publisher-loc>S&#227;o Paulo, SP, Brazil</publisher-loc>
      </publisher>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v3">hRMgtGNmB78J6y4Dqw54YSm</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v2">S0066-782X2012000200008</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v1">S0066-782X(12)09800208</article-id>
      <article-id pub-id-type="doi">10.1590/S0066-782X2012005000013</article-id>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Artigos Originais</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>Avalia&#231;&#227;o da acur&#225;cia do rastre&#244;metro: um novo equipamento para rastreamento de hipertens&#227;o arterial</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <name>
            <surname>Bitencourt</surname>
            <given-names>Fernando</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Gottschall</surname>
            <given-names>Carlos Antonio Mascia</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
        </contrib>
      </contrib-group>
      <aff id="aff01">
        <addr-line>
          <named-content content-type="city">Porto Alegre</named-content>
          <named-content content-type="state">Rio Grande do Sul</named-content>
        </addr-line>
        <institution content-type="orgname">Funda&#231;&#227;o Universit&#225;ria de Cardiologia</institution>
        <institution content-type="orgdiv1">Instituto de Cardiologia do Rio Grande do Sul </institution>
        <country country="BR">Brazil</country>
      </aff>
      <pub-date date-type="collection" publication-format="electronic">
        <month>02</month>
        <year>2012</year>
      </pub-date>
      <volume>98</volume>
      <issue>2</issue>
      <fpage>151</fpage>
      <lpage>160</lpage>
      <history>
        <date date-type="received">
          <day>01</day>
          <month>04</month>
          <year>2011</year>
        </date>
        <date date-type="accepted">
          <day>14</day>
          <month>10</month>
          <year>2011</year>
        </date>
        <date date-type="rev-recd">
          <day>07</day>
          <month>07</month>
          <year>2011</year>
        </date>
      </history>
      <permissions>
        <license license-type="open-access" xlink:href="http://creativecommons.org/licenses/by-nc/4.0/" xml:lang="en">
          <license-p>This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.</license-p>
        </license>
      </permissions>
      <abstract>
        <p>Fundamento: A hipertens&#227;o arterial sist&#234;mica &#233; um dos principais fatores de risco cardiovascular afetando mais de um bilh&#227;o de pessoas. A baixa taxa de conhecimento da doen&#231;a constitui uma das maiores dificuldades relacionadas &#224; hipertens&#227;o. O rastreamento populacional para hipertens&#227;o &#233; amplamente reconhecido como estrat&#233;gia efetiva na detec&#231;&#227;o de casos novos, por&#233;m, at&#233; o momento, n&#227;o se disp&#245;e de nenhum instrumento que torne vi&#225;vel o rastreamento em massa da popula&#231;&#227;o. Objetivo: Avaliar a acur&#225;cia e a reprodutibilidade do Rastre&#244;metro, um novo instrumento para o rastreamento de hipertens&#227;o arterial para agentes comunit&#225;rios de sa&#250;de. M&#233;todos: Estudo transversal com amostra de 436 indiv&#237;duos atendidos pela Estrat&#233;gia de Sa&#250;de da Fam&#237;lia. A avalia&#231;&#227;o foi feita comparando-se os resultados de press&#227;o arterial obtidos pelo m&#233;todo oscilom&#233;trico e a leitura do Rastre&#244;metro realizada pelos agentes comunit&#225;rios de sa&#250;de. Resultados: o Rastre&#244;metro apresentou sensibilidade de 81,3% e especificidade de 85,6% quando avaliado em rela&#231;&#227;o aos valores de press&#227;o sist&#243;lica. O desempenho dos operadores influenciou os resultados do Rastre&#244;metro de forma independente (p = 0,0001). O Rastre&#244;metro apresentou reprodutibilidade substancial (Kendall&#8217;s W 0,71). Conclus&#227;o: O Rastre&#244;metro, quando operado por agentes comunit&#225;rios de sa&#250;de, apresentou boa sensibilidade, especificidade e reprodutibilidade, especialmente quando avaliados &#224; luz dos valores de press&#227;o sist&#243;lica, tendendo a confirmar esse novo instrumento como m&#233;todo de rastreamento para hipertens&#227;o arterial na aten&#231;&#227;o prim&#225;ria a sa&#250;de.</p>
      </abstract>
      <kwd-group xml:lang="pt" kwd-group-type="author-generated">
        <kwd>Hipertens&#227;o</kwd>
        <kwd>press&#227;o arterial</kwd>
        <kwd>equipamentos de medi&#231;&#227;o</kwd>
        <kwd>esfigmomanometros</kwd>
        <kwd>fatores de risco</kwd>
        <kwd>diagn&#243;stico</kwd>
      </kwd-group>
      <pub-date publication-format="electronic" date-type="pub">
        <day>31</day>
        <month>01</month>
        <year>2012</year>
      </pub-date>
    </article-meta>
  </front>
  <body>
    <p>
      <bold>Avalia&#231;&#227;o da acur&#225;cia do rastre&#244;metro: um novo equipamento para rastreamento de hipertens&#227;o arterial</bold>
    </p>
    <p>
      <bold>Fernando Bitencourt; Carlos Antonio Mascia Gottschall</bold>
    </p>
    <p> Instituto de Cardiologia do Rio Grande do Sul/ Funda&#231;&#227;o Universit&#225;ria de Cardiologia - IC/FUC, Porto Alegre, RS, Brasil</p>
    <p>
      <xref ref-type="corresp" rid="end">Correspond&#234;ncia</xref>
    </p>
    <p>
      <bold>RESUMO</bold>
    </p>
    <p><bold>Fundamento</bold>: A hipertens&#227;o arterial sist&#234;mica &#233; um dos principais fatores de risco cardiovascular afetando mais de um bilh&#227;o de pessoas. A baixa taxa de conhecimento da doen&#231;a constitui uma das maiores dificuldades relacionadas &#224; hipertens&#227;o. O rastreamento populacional para hipertens&#227;o &#233; amplamente reconhecido como estrat&#233;gia efetiva na detec&#231;&#227;o de casos novos, por&#233;m, at&#233; o momento, n&#227;o se disp&#245;e de nenhum instrumento que torne vi&#225;vel o rastreamento em massa da popula&#231;&#227;o. </p>
    <p><bold>Objetivo</bold>: Avaliar a acur&#225;cia e a reprodutibilidade do Rastre&#244;metro, um novo instrumento para o rastreamento de hipertens&#227;o arterial para agentes comunit&#225;rios de sa&#250;de.</p>
    <p><bold>M&#233;todos</bold>: Estudo transversal com amostra de 436 indiv&#237;duos atendidos pela Estrat&#233;gia de Sa&#250;de da Fam&#237;lia. A avalia&#231;&#227;o foi feita comparando-se os resultados de press&#227;o arterial obtidos pelo m&#233;todo oscilom&#233;trico e a leitura do Rastre&#244;metro realizada pelos agentes comunit&#225;rios de sa&#250;de.</p>
    <p><bold>Resultados</bold>: o Rastre&#244;metro apresentou sensibilidade de 81,3% e especificidade de 85,6% quando avaliado em rela&#231;&#227;o aos valores de press&#227;o sist&#243;lica. O desempenho dos operadores influenciou os resultados do Rastre&#244;metro de forma independente (p = 0,0001). O Rastre&#244;metro apresentou reprodutibilidade substancial (Kendall&#146;s W 0,71). </p>
    <p><bold>Conclus&#227;o</bold>: O Rastre&#244;metro, quando operado por agentes comunit&#225;rios de sa&#250;de, apresentou boa sensibilidade, especificidade e reprodutibilidade, especialmente quando avaliados &#224; luz dos valores de press&#227;o sist&#243;lica, tendendo a confirmar esse novo instrumento como m&#233;todo de rastreamento para hipertens&#227;o arterial na aten&#231;&#227;o prim&#225;ria a sa&#250;de.</p>
    <p><bold>Palavras-chave</bold>: Hipertens&#227;o/preven&#231;&#227;o, press&#227;o arterial, equipamentos de medi&#231;&#227;o, esfigmomanometros, fatores de risco, diagn&#243;stico.</p>
    <p>
      <bold>Introdu&#231;&#227;o</bold>
    </p>
    <p>A Hipertens&#227;o Arterial Sist&#234;mica (HAS) &#233; um dos principais fatores de risco cardiovascular e principal fator de risco atribu&#237;do para mortalidade global<sup>1</sup>. A detec&#231;&#227;o precoce &#233; primordial no tratamento da hipertens&#227;o visando reduzir eventos cardiovasculares<sup>2</sup>. Por&#233;m, as baixas taxas de conhecimento da doen&#231;a &#233; uma das maiores dificuldades no seu manejo<sup>3,4</sup>. No Brasil, estima-se que metade dos hipertensos conhece sua condi&#231;&#227;o<sup>5</sup>. O rastreamento populacional para hipertens&#227;o &#233; reconhecido como estrat&#233;gia efetiva na detec&#231;&#227;o de casos novos<sup>6,7</sup>. </p>
    <p>At&#233; o momento, no entanto, n&#227;o se disp&#245;e de nenhum instrumento que torne vi&#225;vel o rastreamento em massa da popula&#231;&#227;o. </p>
    <p>Estudos<sup>8,9</sup> t&#234;m demonstrado que um novo instrumento, o Rastre&#244;metro, poderia ser uma possibilidade para o rastreamento populacional para HAS no &#226;mbito da aten&#231;&#227;o prim&#225;ria, uma vez que foi projetado para ser utilizado por Agentes Comunit&#225;rios de Sa&#250;de (ACS). Isso significa que os mais de 240 mil ACS<sup>10 </sup>espalhados pelo pa&#237;s poderiam aplicar esse teste de forma r&#225;pida, com custos baixos e riscos m&#237;nimos, para detectar suspeitos de hipertens&#227;o No entanto, o Rastre&#244;metro foi testado somente por profissionais e estudantes da &#225;rea m&#233;dica e em ambientes controlados<sup>8,9</sup>.</p>
    <p>O Rastre&#244;metro (<xref ref-type="fig" rid="a08fig01_">fig. 1</xref>) &#233; uma adapta&#231;&#227;o de um esfigmoman&#244;metro aneroide em que o visor do man&#244;metro &#233; coberto por um adesivo com duas faixas, uma amarela indicando valores de press&#227;o &lt; 140 mmHg e outra vermelha indicando valores = ou &gt; a 140 mmHg. A avalia&#231;&#227;o &#233; feita pela leitura do in&#237;cio da oscila&#231;&#227;o da agulha do man&#244;metro, sendo considerado suspeito de HAS aquele em que a oscila&#231;&#227;o da agulha iniciar na faixa vermelha, e normotenso aqueles com in&#237;cio da oscila&#231;&#227;o na faixa amarela. Uma limita&#231;&#227;o dessa t&#233;cnica &#233; a impossibilidade, <italic>a priori</italic>, de avaliar altera&#231;&#245;es isoladas da press&#227;o diast&#243;lica.</p>
    <p content-type="asset">
      <fig id="a08fig01_">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gf01.jpg"/></fig>
    </p>
    <p>Essa t&#233;cnica foi desenvolvida em raz&#227;o da proibi&#231;&#227;o, pelo Conselho Federal de Enfermagem, de os ACS realizarem medidas de press&#227;o arterial pelo m&#233;todo auscultat&#243;rio/oscilom&#233;trico.</p>
    <p>O objetivo deste trabalho &#233; avaliar a acur&#225;cia e a reprodutibilidade do Rastre&#244;metro como m&#233;todo de rastreamento para HAS quando operado por agentes comunit&#225;rios de sa&#250;de.</p>
    <p>
      <bold>M&#233;todos</bold>
    </p>
    <p>Estudo transversal realizado na cidade de Charqueadas (RS). A popula&#231;&#227;o do estudo foram os habitantes adscritos da Unidade de Sa&#250;de da Fam&#237;lia Sul Am&#233;rica. Os participantes da pesquisa foram selecionados por amostragem probabil&#237;stica aleat&#243;ria simples.</p>
    <p>O c&#225;lculo amostral para avalia&#231;&#227;o da acur&#225;cia do Rastre&#244;metro foi realizado baseando-se nos valores de sensibilidade (95%) e especificidade (63,1%) apresentados nos estudos anteriores e sobre uma preval&#234;ncia de hipertens&#227;o na &#225;rea do estudo (21%). Foram admitidos uma margem de erro de 5% e poder de 80%. O resultado do c&#225;lculo amostral foi 436 pessoas. </p>
    <p>A fim de testar a reprodutibilidade do Rastre&#244;metro, uma segunda amostra de 90 pacientes foi avaliada. Para esse teste, utilizou-se amostra de conveni&#234;ncia de pessoas que procuraram atendimento em uma segunda Unidade de Sa&#250;de do munic&#237;pio.</p>
    <p>Foram considerados crit&#233;rios de inclus&#227;o: circunfer&#234;ncia braquial entre 22 cm e 42 cm, uso regular da terapia farmacol&#243;gica anti-hipertensiva nos pacientes com diagn&#243;stico de hipertens&#227;o e idade entre 20 e 80 anos. </p>
    <p>O m&#233;todo Rastre&#244;metro foi comparado com o m&#233;todo oscilom&#233;trico, considerado padr&#227;o de refer&#234;ncia no estudo. O Rastre&#244;metro foi confeccionado a partir de um esfigmoman&#244;metro aneroide da marca <italic>Eternum</italic> tamanho adulto (bolsa de borracha de 12 cm de largura e 23 cm de comprimento), indicado para circunfer&#234;ncias braquiais entre 24 e 30 cm. Como padr&#227;o de refer&#234;ncia, foi utilizado esfigmoman&#244;metro digital oscilom&#233;trico da marca<italic> Microlife</italic> modelo 3BTO-A, validado pela <italic>British Hypertension Society</italic><sup>11</sup>. Nesse aparelho, foram usados dois tipos de bra&#231;adeiras com tamanhos apropriados para cada circunfer&#234;ncia braquial (<italic>Standard adult</italic> 22-32cm e <italic>Large adult</italic> 32-42cm). Esse aparelho digital foi usado como compara&#231;&#227;o na tentativa de evitar vieses de aferi&#231;&#227;o<sup>12,13</sup>.</p>
    <p>Previamente ao in&#237;cio do trabalho, os instrumentos da pesquisa, que eram novos e certificados pelo Instituto Nacional de Metrologia, Normaliza&#231;&#227;o e Qualidade Industrial &#150; Inmetro &#150; (com selo de certifica&#231;&#227;o de Maio de 2008), foram avaliados a fim de averiguar adequabilidade de calibra&#231;&#227;o entre os esfigmoman&#244;metros. Ap&#243;s o in&#237;cio da pesquisa foram realizadas mais duas verifica&#231;&#245;es de calibragem. As avalia&#231;&#245;es de calibragem foram realizadas pelo <italic>Inmetro</italic>. Em nenhuma das avalia&#231;&#245;es houve discrep&#226;ncia em rela&#231;&#227;o &#224; calibra&#231;&#227;o entre os instrumentos ou reprova&#231;&#227;o de qualquer outro item avaliado. </p>
    <p>O Rastre&#244;metro foi utilizado em participantes com circunfer&#234;ncia braquial entre 22 cm e 42 cm. Para a avalia&#231;&#227;o da press&#227;o arterial foi utilizada bra&#231;adeira espec&#237;fica para as diferentes circunfer&#234;ncias braquiais, seguindo as especifica&#231;&#245;es do fabricante do esfigmoman&#244;metro <italic>Microlife</italic>. Essa op&#231;&#227;o por comparar um valor de press&#227;o arterial mais pr&#243;ximo do valor real (pela utiliza&#231;&#227;o de bra&#231;adeiras apropriadas para circunfer&#234;ncia braquial) com o obtido por meio da leitura do Rastre&#244;metro (somente com aparelho tipo adulto) serviu para avaliar a influ&#234;ncia da circunfer&#234;ncia braquial na acur&#225;cia do Rastre&#244;metro, visto que o tamanho do manguito exerce influ&#234;ncia inquestion&#225;vel nas medidas da press&#227;o arterial pelos m&#233;todos indiretos<sup>14</sup>.</p>
    <p>O m&#233;todo Rastre&#244;metro foi manuseado por quatro ACS e suas leituras foram comparadas com o m&#233;todo oscilom&#233;trico, cujas mensura&#231;&#245;es foram realizadas por um enfermeiro<sup>15</sup>.</p>
    <p>Todos os ACS utilizaram o mesmo aparelho em todas as fases da pesquisa. Os ACS n&#227;o conheciam o novo m&#233;todo; assim, receberam um treinamento b&#225;sico sobre a t&#233;cnica de leitura do Rastre&#244;metro e, ap&#243;s, cada um realizou 30 medidas com o instrumento em pacientes diferentes. </p>
    <p>Inicialmente, foram aplicados os question&#225;rios e obtidas as medidas antropom&#233;tricas para cada participante. A entrevista, as medidas antropom&#233;tricas e a avalia&#231;&#227;o da press&#227;o arterial pelo m&#233;todo oscilom&#233;trico ocorreram somente com a presen&#231;a do enfermeiro, a fim de manter os ACS cegos em rela&#231;&#227;o &#224;s caracter&#237;sticas cl&#237;nicas e aos valores de press&#227;o arterial. Os operadores do Rastre&#244;metro e do aparelho padr&#227;o de refer&#234;ncia tamb&#233;m ficaram cegos entre si. Cada participante teve sua press&#227;o arterial medida duas vezes pelo m&#233;todo oscilom&#233;trico e duas vezes com o Rastre&#244;metro. Primeiramente, foram realizadas as avalia&#231;&#245;es pelo m&#233;todo oscilom&#233;trico e, ap&#243;s, com o Rastre&#244;metro. Todas as avalia&#231;&#245;es da press&#227;o arterial seguiram as orienta&#231;&#227;o das V Diretrizes Brasileiras de Hipertens&#227;o Arterial<sup>2</sup>.As avalia&#231;&#245;es foram realizadas ap&#243;s um repouso inicial de 5 minutos e com intervalo de 1 minuto entre a primeira e a segunda medidas (m&#233;todo padr&#227;o de refer&#234;ncia) e entre a terceira e a quarta medidas (Rastre&#244;metro). Entre a segunda e a terceira medidas foi respeitado um intervalo de 2 minutos<sup>2</sup>. </p>
    <p>Considerou-se positivo para press&#227;o arterial elevada com o Rastre&#244;metro se uma ou ambas as leituras realizadas fossem positivas (oscila&#231;&#227;o da agulha na faixa vermelha indicando press&#227;o arterial sist&#243;lica igual ou maior que 140 mmHg). Para a compara&#231;&#227;o entre os resultados, foi considerada a m&#233;dia aritm&#233;tica dos valores de press&#227;o arterial obtidos com o m&#233;todo oscilom&#233;trico. Foram considerados como press&#227;o arterial elevada valores iguais ou superiores a 140 mmHg para press&#227;o sist&#243;lica e 90 mmHg para a press&#227;o diast&#243;lica<sup>2</sup>. O uso da regular da terapia farmacol&#243;gica foi avaliada pelo question&#225;rio de Morisky-Green<sup>16</sup>.</p>
    <p>Para a avalia&#231;&#227;o da reprodutibilidade do Rastre&#244;metro, foi realizada uma nova sequ&#234;ncia de avalia&#231;&#245;es em uma amostra independente da primeira. Nesse teste, cada ACS realizou uma medida com o Rastre&#244;metro com intervalo de um minuto entre cada medida em um mesmo participante. Nessa avalia&#231;&#227;o tamb&#233;m foi obtida uma medida da press&#227;o arterial pelo m&#233;todo oscilom&#233;trico. Durante os testes de reprodutibilidade, os ACS ficaram cegos entre si, para o diagnostico pr&#233;vio de hipertens&#227;o e valor da press&#227;o arterial.</p>
    <p>Foram utilizados teste <italic>t de Student </italic>para a compara&#231;&#227;o de vari&#225;veis cont&#237;nuas e teste Qui-quadrado para vari&#225;veis categ&#243;ricas. Para identificar as vari&#225;veis que influenciaram de forma independente os resultados do Rastre&#244;metro foram realizadas an&#225;lises bivari&#225;vel e multivari&#225;vel. Primeiramente, todas as vari&#225;veis de interesse foram testadas em modelo bivariado, como segue: idade, sexo, ra&#231;a, &#205;ndice de Massa Corp&#243;rea (IMC &gt; 30 / IMC &lt; 30), frequ&#234;ncia card&#237;aca, tamanho do manguito utilizado no Rastre&#244;metro (adequado / n&#227;o adequado para circunfer&#234;ncia braquial), diagn&#243;stico pr&#233;vio de HAS, diabetes ou dislipidemia, h&#225;bito tab&#225;gico e operador (compara&#231;&#227;o do percentual de acerto entre os operadores).</p>
    <p>Para as avalia&#231;&#245;es de acur&#225;cia foram realizadas an&#225;lises de sensibilidade, especificidade, valor preditivo positivo e negativo, probabilidade pr&#233;-teste, raz&#227;o de verossimilhan&#231;a<italic> +</italic>, raz&#227;o de verossimilhan&#231;a<italic> &#150;</italic>, probabilidade p&#243;s-teste + e probabilidade p&#243;s-teste<sup>17</sup>. Para avaliar a concord&#226;ncia intraobservador foi utilizado teste de Kappa, e para a avalia&#231;&#227;o da concord&#226;ncia entre os observadores foi utilizado teste de Kendall&#146;s W. Foi considerado como signific&#226;ncia estat&#237;stica um valor p &lt; 0,05 e utilizado um intervalo de confian&#231;a de 95%. O poder dos testes estat&#237;sticos foi de 80%.</p>
    <p>O protocolo foi aprovado pelo Comit&#234; de &#201;tica em Pesquisa e todos os participantes assinaram o termo de consentimento livre e esclarecido.</p>
    <p>
      <bold>Resultados</bold>
    </p>
    <p>A primeira visita aconteceu no dia 23/8/2008 e a &#250;ltima, no dia 24/7/2009. O fluxograma do estudo &#233; apresentado na figura 2<sup>18</sup>.</p>
    <p>Na amostra houve predom&#237;nio do sexo feminino (61,9%) e da cor branca (77,8%). A m&#233;dia de idade foi 47 anos. O n&#250;mero total de pacientes com diagn&#243;stico pr&#233;vio de HAS foi de 131, sendo o percentual de pacientes hipertensos com a press&#227;o controlada de 58%. Oitenta e um pacientes estavam com a Press&#227;o Arterial (PA) elevada no momento do teste. As demais caracter&#237;sticas cl&#237;nicas e demogr&#225;ficas da amostra s&#227;o apresentadas na <xref ref-type="table" rid="a08tab01_">tabela 1</xref>.</p>
    <p content-type="asset">
      <table-wrap id="a08tab01_">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt01.jpg"/></table-wrap>
    </p>
    <p>Uma vez que o Rastre&#244;metro tem a capacidade, <italic>a priori</italic>, de identificar altera&#231;&#245;es do componente sist&#243;lico da press&#227;o arterial, possibilitando somente o rastreamento de casos de hipertens&#227;o sist&#243;lica isolada e sist&#243;lica / diast&#243;lica, os resultados descritos a seguir ser&#227;o apresentados, primeiro considerando hipertens&#227;o arterial conforme os valores de refer&#234;ncia da VI Diretrizes Brasileira de Hipertens&#227;o Arterial<sup>5</sup>, e ser&#225; descrita como Avalia&#231;&#227;o da Press&#227;o Sist&#243;lica e Diast&#243;lica (APSD); segundo, considerando somente o valor de press&#227;o sist&#243;lica igual ou maior que 140 mmHg, caracterizando a HAS, descartando os casos de hipertens&#227;o diast&#243;lica isolada, e ser&#225; descrita como Avalia&#231;&#227;o da Press&#227;o Sist&#243;lica (APS).</p>
    <p>Os resultados para APSD e APS s&#227;o apresentados na <xref ref-type="table" rid="a08tab02_">tabela 2</xref>. Com a avalia&#231;&#227;o isolada dos valores da press&#227;o sist&#243;lica houve aumento de 10,8% no valor de sensibilidade e discreta redu&#231;&#227;o da especificidade (1,7%). Em rela&#231;&#227;o aos valores preditivos, houve redu&#231;&#227;o no valor preditivo positivo (8,3%), j&#225; esperada em raz&#227;o da redu&#231;&#227;o da preval&#234;ncia, e discreto aumento do valor preditivo negativo (3,9%). No entanto, as raz&#245;es de verossimilhan&#231;a positivas e negativas foram apenas marginalmente alteradas, com discreto aumento em APS.</p>
    <p content-type="asset">
      <table-wrap id="a08tab02_">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt02.jpg"/></table-wrap>
    </p>
    <p>O desempenho de cada ACS tamb&#233;m foi avaliado considerando APS e APSD. Os resultados s&#227;o apresentados nas <xref ref-type="table" rid="a08tab03_">tabelas 3</xref> e <xref ref-type="table" rid="a08tab04_">4</xref>. Nota-se que houve uma melhora nos resultados de sensibilidade e discreta redu&#231;&#227;o na especificidade tamb&#233;m nos resultados individuais dos operadores quando considerada a APS. As diferen&#231;as entre os agentes, no entanto, persistiram nas duas avalia&#231;&#245;es. </p>
    <p content-type="asset">
      <table-wrap id="a08tab03_">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt03.jpg"/></table-wrap>
    </p>
    <p content-type="asset">
      <table-wrap id="a08tab04_">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt04.jpg"/></table-wrap>
    </p>
    <p>O n&#250;mero de casos de hipertens&#227;o diast&#243;lica isolada (22 pacientes tiveram eleva&#231;&#227;o isolada da press&#227;o diast&#243;lica) e a interpreta&#231;&#227;o de cada ACS s&#227;o apresentados na <xref ref-type="table" rid="t05">tabela 5</xref>.</p>
    <table-wrap id="t05">
      <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt05.jpg"/></table-wrap>
    <p>A reprodutibilidade do Rastre&#244;metro foi avaliada mediante a concord&#226;ncia interobservador pelo teste de Kendall&#146;s W com resultado de 0,71. Medidas de concord&#226;ncia intraobservador tamb&#233;m foram realizadas para cada ACS por meio do teste de Kappa (tab. 3 e 4). Uma medida geral de concord&#226;ncia foi realizada para avaliar a concord&#226;ncia geral do Rastre&#244;metro com o m&#233;todo padr&#227;o de refer&#234;ncia obtendo um valor Kappa de 0,52 (IC 95% 0,43 - 0,62) para APSD e Kappa de 0,51(IC 95% 0,40 - 0,62) para APS. </p>
    <p>Tamb&#233;m foram geradas curvas de aprendizado para os agentes de sa&#250;de, a fim de avaliar o desempenho de cada operador ao longo do estudo. Foi realizada a compara&#231;&#227;o entre os percentuais de acerto entre quintis das amostras de cada ACS. Essas evidenciaram que nenhum dos operadores variou seu desempenho ao decorrer das avalia&#231;&#245;es (p = ns).</p>
    <p>Na an&#225;lise de regress&#227;o de Poisson, a &#250;nica vari&#225;vel que influenciou de forma independente os resultados do Rastre&#244;metro foi o operador (p = 0,0001). O uso de bra&#231;adeira do esfigmoman&#244;metro inadequado para a circunfer&#234;ncia braquial (p = 0,19) e o diagn&#243;stico pr&#233;vio de HAS (p = 0,10) n&#227;o influenciaram os resultados do Rastre&#244;metro de forma significativa.</p>
    <p>
      <bold>Discuss&#227;o</bold>
    </p>
    <p>Este foi o primeiro trabalho com o Rastre&#244;metro envolvendo agentes comunit&#225;rios de sa&#250;de para avalia&#231;&#227;o de sua acur&#225;cia como teste de rastreamento para hipertens&#227;o arterial de que temos conhecimento. Neste estudo, o Rastre&#244;metro apresentou boa sensibilidade e especificidade, especialmente quando avaliado em rela&#231;&#227;o aos valores de press&#227;o sist&#243;lica. O uso de bra&#231;adeira inadequada para a circunfer&#234;ncia braquial no Rastre&#244;metro (p = 0,19) e o diagn&#243;stico pr&#233;vio de HAS (p = 0,10) n&#227;o influenciaram os resultados do novo instrumento. O desempenho dos operadores (ACS) influenciou os resultados do Rastre&#244;metro de forma independente (p = 0,0001).  O Rastre&#244;metro apresentou reprodutibilidade substancial quando avaliada pelo teste de Kendall&#146;s W (0,71).</p>
    <p>Este foi o terceiro estudo realizado envolvendo esse novo m&#233;todo para rastreamento de HAS; no entanto, foi o primeiro a avaliar a acur&#225;cia do Rastre&#244;metro com ACS e em um ambiente de aten&#231;&#227;o prim&#225;ria. </p>
    <p>Nesta avalia&#231;&#227;o o Rastre&#244;metro apresentou sensibilidade de 70,5% (IC 95% 57,0 - 79,7) e especificidade de 87,3% (IC 95% 83,3 - 90,5) quando avaliado em rela&#231;&#227;o aos valores de press&#227;o sist&#243;lica e diast&#243;lica. Considerando apenas os valores de press&#227;o sist&#243;lica o Rastre&#244;metro aumentou sua sensibilidade para 81,3% (IC 95% 68,6-89,8) e teve uma pequena redu&#231;&#227;o de sua especificidade que se estabeleceu em 85,6% (IC 95% 81,6 - 88,9). </p>
    <p>Os achados deste estudo corroboram os resultados dos estudos anteriores. No primeiro estudo<sup>8</sup>, o Rastre&#244;metro foi testado pelo seu pr&#243;prio inventor em uma amostra de 200 pacientes. Dentre esses, 100 eram sabidamente sem diagn&#243;stico de hipertens&#227;o e 100 sabidamente hipertensos. Os resultados de sensibilidade e de especificidade para o grupo sem diagn&#243;stico de hipertens&#227;o foram bastante altos, chegando a valores de 96,7% para sensibilidade e 80% para especificidade. J&#225; no grupo com diagn&#243;stico da doen&#231;a, a sensibilidade foi de 98,7% e especificidade de 94,7%. No segundo estudo<sup>9</sup>, o Rastre&#244;metro foi testado em uma amostra de 268 pacientes, por dois estudantes de medicina, com resultados de 95,1% para sensibilidade e de 63,1% para especificidade.</p>
    <p>Embora os resultados dos tr&#234;s estudos tendam a confirmar o Rastre&#244;metro como um m&#233;todo de triagem populacional para HAS, algumas considera&#231;&#245;es s&#227;o pertinentes. </p>
    <p>Ambos os estudos anteriores foram realizados em ambientes de aten&#231;&#227;o secund&#225;ria. Dessa forma, os resultados podem ter sido superestimados pelas caracter&#237;sticas das popula&#231;&#245;es estudadas, a saber, maior preval&#234;ncia de HAS e o espectro da doen&#231;a dos pacientes nesses estudos<sup>17</sup>. Por si s&#243;s, esses dois fatores poderiam aumentar a sensibilidade, em raz&#227;o do vi&#233;s de espectro, e de valor preditivo positivo, relacionado a maior preval&#234;ncia de doen&#231;a<sup>19</sup>. </p>
    <p>A falta de cegamento dos operadores do Rastre&#244;metro para as caracter&#237;sticas cl&#237;nicas dos participantes nos estudos pr&#233;vios tamb&#233;m poderia estar relacionada a aumento da sensibilidade do instrumento em teste, visto que o conhecimento do diagn&#243;stico do paciente antes da aplica&#231;&#227;o do teste pode levar, mesmo que inconscientemente, a identificar como doentes pacientes com diagn&#243;stico pr&#233;vio estabelecido<sup>17,19</sup>. Outro fato importante foi o n&#250;mero de pacientes em que o Rastre&#244;metro foi aplicado nos estudos anteriores, o que poderia acarretar em vi&#233;s de acaso<sup>19</sup>. </p>
    <p>Vale destacar que as caracter&#237;sticas dos operadores dos tr&#234;s estudos s&#227;o razoavelmente distintas. O Rastre&#244;metro foi operado por um cardiologista experiente e inventor do instrumento no primeiro estudo, por dois estudantes de medicina no segundo e, finalmente, por quatro ACS com o ensino m&#233;dio completo. Esse fato tamb&#233;m poderia acarretar em diferen&#231;a entre os resultados dos estudos a favor dos primeiros supracitados<sup>17</sup>. Outro fator importante foi a precis&#227;o diagn&#243;stica utilizada nos diferentes estudos. No estudo de Forsvall<sup>9</sup> foi utilizado padr&#227;o de refer&#234;ncia com precis&#227;o de 5 mmHg. Em nosso estudo, a precis&#227;o nas medidas da press&#227;o arterial foi de 1 mmHg. A precis&#227;o das medidas da PA no primeiro estudo n&#227;o foi relatada.</p>
    <p>Dessa forma, o desempenho do Rastre&#244;metro neste estudo &#233; compat&#237;vel com o descrito por Fletcher<sup>19</sup>, que refor&#231;a que por ser aplicado em uma popula&#231;&#227;o assintom&#225;tica e com casos mais leves de doen&#231;a, os testes de rastreamento tendem a ter uma sensibilidade mais baixa e uma especificidade mais alta do que quando aplicado em um espectro de pacientes mais severo; e que o valor preditivo positivo &#233; geralmente baixo, mesmo com altos valores de especificidade. </p>
    <p>A avalia&#231;&#227;o realizada em rela&#231;&#227;o aos valores isolados da PA sist&#243;lica &#233; bastante pertinente, visto que o Rastre&#244;metro tem uma limita&#231;&#227;o intr&#237;nseca, que &#233; a capacidade de avaliar somente o marcador de press&#227;o sist&#243;lica (in&#237;cio da oscila&#231;&#227;o da agulha). A hipertens&#227;o sist&#243;lica, no entanto, &#233; a forma mais prevalente de HAS e o mais importante fator de risco cardiovascular<sup>20</sup>. N&#227;o obstante, o valor isolada do componente sist&#243;lico da press&#227;o arterial &#233; capaz de identificar corretamente 91% dos casos de hipertens&#227;o arterial<sup>21</sup>. </p>
    <p>Nesse estudo, 24 pacientes com altera&#231;&#227;o da press&#227;o arterial n&#227;o foram identificados pelo Rastre&#244;metro (<xref ref-type="fig" rid="a08fig02_">fig. 2</xref>), e desses, 11 apresentavam eleva&#231;&#227;o da PA sist&#243;lica e 13 apresentavam eleva&#231;&#227;o isolada da PA diast&#243;lica. Entretanto, o Rastre&#244;metro identificou nove pacientes com altera&#231;&#227;o isolada da PA diast&#243;lica, de um total de 22 casos, o que n&#227;o era esperado que ocorresse (tab. 5). &#192; primeira vista, a n&#227;o identifica&#231;&#227;o de 24 suspeitos de HAS poderia ser considerada expressiva; no entanto, se considerarmos o contexto em que se pretende usar o instrumento, a estrat&#233;gia de sa&#250;de da fam&#237;lia, entenderemos que os casos n&#227;o rastreados no primeiro momento poder&#227;o ser identificados nas triagens seguintes, visto que o teste poder&#225; ser aplicado mensalmente pelos agentes comunit&#225;rios.</p>
    <p content-type="asset">
      <fig id="a08fig02_">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gf02.jpg"/></fig>
    </p>
    <p>Por sua vez, 45 pacientes (<xref ref-type="fig" rid="a08fig02_">fig. 2</xref>) foram identificados como falso-positivos pelo Rastre&#244;metro. Destaca-se que a identifica&#231;&#227;o de casos falso-positivos teria uma consequ&#234;ncia benigna, que seria a ida &#224; Unidade de Sa&#250;de para avalia&#231;&#227;o da press&#227;o arterial e, caso necess&#225;rio, esclarecimento diagn&#243;stico em consulta m&#233;dica. </p>
    <p>O uso de um &#250;nico tamanho de manguito, independentemente da circunfer&#234;ncia braquial, n&#227;o afetou de forma independente as propriedades do Rastre&#244;metro (p = 0,19). Redu&#231;&#227;o na acur&#225;cia do instrumento em situa&#231;&#245;es de inadequa&#231;&#227;o do manguito era esperada, visto que a influ&#234;ncia do tamanho do manguito na medida da press&#227;o arterial j&#225; &#233; bem definida<sup>12,14,22</sup>. No entanto, essa situa&#231;&#227;o pode ser explicada em partes, pelo fato de ter sido utilizado um manguito menor que o indicado no Rastre&#244;metro em 80,7% da amostra, o que aumentou de certa forma sua sensibilidade devido a uma leitura superestimada da press&#227;o arterial<sup>14 </sup>e n&#227;o influenciou a especificidade. Esse dado confirma o achado de Forsvall<sup>9</sup>, que tamb&#233;m n&#227;o identificou diferen&#231;as na precis&#227;o do Rastre&#244;metro quando utilizado manguito inadequado para a circunfer&#234;ncia do bra&#231;o; no entanto, naquele estudo foram utilizados manguitos id&#234;nticos no Rastre&#244;metro e no padr&#227;o de refer&#234;ncia. </p>
    <p>Vale destacar que a avalia&#231;&#227;o da influ&#234;ncia da circunfer&#234;ncia braquial na acur&#225;cia do Rastre&#244;metro &#233; imposs&#237;vel de ser realizada se usados manguitos inadequados para a circunfer&#234;ncia braquial no instrumento padr&#227;o de refer&#234;ncia, pois o resultado obtido nesse aparelho n&#227;o seria fidedigno &#224; realidade, mas compat&#237;vel com as medidas do Rastre&#244;metro. O fato de uma poss&#237;vel superestima&#231;&#227;o dos valores de sensibilidade no Rastre&#244;metro, pela utiliza&#231;&#227;o de manguito menor que o exigido no Rastre&#244;metro, de longe &#233; negativo, uma vez que se esperam altos valores de sensibilidade em testes de rastreamento. </p>
    <p>Em nosso estudo, o diagn&#243;stico pr&#233;vio de HAS tamb&#233;m n&#227;o influenciou de forma significativa as propriedades do Rastre&#244;metro (p = 0,10). Entretanto, no trabalho de Forsvall<sup>9</sup>, quando comparadas as leituras alteradas do Rastre&#244;metro com resultados de press&#227;o arterial acima de 140 mmHg de PA sist&#243;lica ou 90 mmHg de PA diast&#243;lica,o uso de medica&#231;&#227;o para tratamento da HAS, o que implica ser hipertenso, alterou a especificidade do Rastre&#244;metro de 77,8% no grupo n&#227;o medicado para 32,7% no grupo medicado. O uso regular da medica&#231;&#227;o anti-hipertensiva foi avaliado em nosso estudo com o objetivo de identificar varia&#231;&#245;es de acur&#225;cia relacionadas ao medicamento, comparando pacientes hipertensos em tratamento farmacol&#243;gico e n&#227;o farmacol&#243;gico isolado. No entanto, n&#227;o houve casos de tratamento n&#227;o farmacol&#243;gico isolado na amostra, o que impossibilitou a avalia&#231;&#227;o da influ&#234;ncia do uso de anti-hipertensivos.</p>
    <p>Um das possibilidades para explicar essa diferen&#231;a entre os estudos foi a falta de cegamento em rela&#231;&#227;o ao diagn&#243;stico do paciente antes das aferi&#231;&#245;es com o Rastre&#244;metro naquele estudo, o que pode ter levado a uma tend&#234;ncia em reconhecer os teste aplicados em pacientes sabidamente hipertensos como alterados, mesmo quando esses estavam com a press&#227;o controlada, incorrendo da&#237; a redu&#231;&#227;o observada no valor de especificidade. </p>
    <p>A explica&#231;&#227;o do autor para justificar aquela diferen&#231;a entre os resultados de especificidade nos grupos medicado e n&#227;o medicado foi atribu&#237;da &#224; falta de padroniza&#231;&#227;o do in&#237;cio das oscila&#231;&#245;es que identificam a real press&#227;o sist&#243;lica e que um melhor treinamento dos operadores poderia corrigir essa distor&#231;&#227;o. Naquele estudo, os operadores do Rastre&#244;metro atingiram seu plat&#244; de desempenho somente ao final da pesquisa. Esse fato n&#227;o &#233; suportado pelos nossos achados, visto que os ACS n&#227;o variaram seus desempenhos de forma significativa durante a coleta dos dados, denotando que os agentes atingiram o plat&#244; de aprendizado ainda na fase de treinamento.</p>
    <p>Por sua vez, a vari&#225;vel operador foi identificada como fator independente na varia&#231;&#227;o dos resultados do Rastre&#244;metro (p = 0,0001), mostrando que os ACS variaram seus desempenhos entre si. Essa marcada diferen&#231;a entre um e outro ACS em rela&#231;&#227;o ao desempenho com o Rastre&#244;metro indica que os resultados do aparelho possam estar ligados &#224;s habilidades de cada operador, visto n&#227;o ter havido diferen&#231;a entre as m&#233;dias de press&#227;o arterial entre as amostras de cada ACS (p &gt; 0,05).</p>
    <p>Outra hip&#243;tese para a grande variabilidade entre os operadores foi o fato de a agulha do man&#244;metro n&#227;o manter um padr&#227;o definido de oscila&#231;&#227;o em todos os pacientes, o que dificulta a identifica&#231;&#227;o do exato momento em que o fluxo sangu&#237;neo come&#231;a a passar pela art&#233;ria e movimentar a agulha do man&#244;metro, o que seria o marcador da press&#227;o sist&#243;lica no Rastre&#244;metro. Essa falta de um padr&#227;o definido de oscila&#231;&#227;o pode ser atribu&#237;da &#224; varia&#231;&#227;o de caracter&#237;sticas end&#243;genas do pr&#243;prio paciente, como volume de sangue ejetado na aorta, contratilidade mioc&#225;rdica e rigidez a&#243;rtica, uma vez que esses tr&#234;s fatores s&#227;o determinantes da press&#227;o arterial sist&#243;lica<sup>23</sup>. A falta de um padr&#227;o definido de oscila&#231;&#227;o j&#225; havia sido comentada por Forsvall<sup>9</sup>. O reduzido n&#250;mero de casos de press&#227;o arterial elevada em cada amostra pode ter acentuado a variabilidade entre os ACS pelo vi&#233;s de acaso<sup>19</sup>. </p>
    <p>Embora tenha havido varia&#231;&#227;o de desempenho entre os ACS quando avaliados em amostras independentes, o teste realizado em uma segunda amostra para avaliar a reprodutibilidade do m&#233;todo apresentou resultado de concord&#226;ncia substancial (Kendall&#146;s W 0,71). O fato de ter havido uma menor discord&#226;ncia entre os operadores quando avaliados os mesmos pacientes em uma amostra de conveni&#234;ncia refor&#231;a a influ&#234;ncia do espectro da doen&#231;a na acur&#225;cia do instrumento, uma vez que houve diferen&#231;a na m&#233;dia da press&#227;o arterial entre as amostras dos testes de acur&#225;cia (122,7 &#177; 17,5) e reprodutibilidade (126,9 &#177; 20,6) com p-valor = 0,04. </p>
    <p>As avalia&#231;&#245;es da press&#227;o arterial pelos m&#233;todos padr&#227;o de refer&#234;ncia e rastre&#244;metro realizadas em sequ&#234;ncia n&#227;o aleat&#243;ria foi uma fonte potencial de vi&#233;s em nosso estudo, pela possibilidade de superestima&#231;&#227;o das primeiras medidas (padr&#227;o de refer&#234;ncia), o que acarretaria em redu&#231;&#227;o da sensibilidade. Todavia, a avalia&#231;&#227;o da PA realizada por enfermeiro pode ter minimizado o efeito do avental branco<sup>15</sup>, uma vez que n&#227;o houve diferen&#231;a entre as m&#233;dias das primeiras e das segundas aferi&#231;&#245;es da press&#227;o arterial (p &gt; 0,05).</p>
    <p>
      <bold>Conclus&#227;o</bold>
    </p>
    <p>O Rastre&#244;metro, quando operado por agentes comunit&#225;rios de sa&#250;de, apresentou boa sensibilidade, especificidade e reprodutibilidade, especialmente quando avaliados sob a luz dos valores de press&#227;o sist&#243;lica, tendendo a confirmar esse novo instrumento como m&#233;todo de rastreamento para HAS na aten&#231;&#227;o prim&#225;ria &#224; sa&#250;de. O Rastre&#244;metro poderia ser utilizado como uma ferramenta no trabalho di&#225;rio dos agentes comunit&#225;rios de sa&#250;de auxiliando na capta&#231;&#227;o precoce de pacientes hipertensos, possibilitando, assim, uma melhoria nos &#237;ndices de conhecimento da doen&#231;a, hoje uma das maiores dificuldades relacionadas &#224; hipertens&#227;o arterial.</p>
    <p>Demais estudos envolvendo agentes de sa&#250;de em outros contextos s&#227;o necess&#225;rios para validar esse novo instrumento. A avalia&#231;&#227;o da efetividade do uso do Rastre&#244;metro tamb&#233;m deve ser investigada futuramente mediante estudos randomizados. </p>
    <p>
      <fn id="end" fn-type="corresp">
        <p>Correspond&#234;ncia:</p>
        <p>Carlos Antonio Mascia Gottschall</p>
        <p>Av. Princesa Isabel, 370 U. Pesquisa, Santana - 90620-000 - Porto Alegre, RS, Brasil</p>
        <p> E-mail: </p>
        <p><ext-link xlink:href="mailto: gottschall.pesquisa@cardiologia.org.br" ext-link-type="email">gottschall.pesquisa@cardiologia.org.br</ext-link>, </p>
        <p>
          <ext-link xlink:href="mailto:'editoracaopc@cardiologia.org.br'" ext-link-type="email">editoracaopc@cardiologia.org.br</ext-link>
        </p>
      </fn>
    </p>
    <p>Artigo recebido em 01/04/11; revisado recebido em 07/07/11; aceito em 14/10/11.</p>
  </body>
  <back>
    <ref-list>
      <ref id="B1">
        <mixed-citation>1. World Health Organization (WHO).   World Health Report 2002: reducing risks, promoting healthy life. Geneva (Switzerland); 2002.</mixed-citation>
        <element-citation publication-type="book">
          <source>World Health Report 2002: reducing risks, promoting healthy life</source>
          <date>
            <year>2002</year>
          </date>
        </element-citation>
      </ref>
      <ref id="B2">
        <mixed-citation>2. Sociedade Brasileira de Cardiologia. Sociedade Brasileira de Hipertens&#259;o. Sociedade Brasileira de Nefrologia. V Diretrizes brasileiras de hipertens&#259;o arterial sist&#281;mica. Arq Bras Cardiol. 2006;89(3):e24-e99.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Sociedade Brasileira de Hipertens&#227;o: Sociedade Brasileira de Nefrologia. V Diretrizes brasileiras de hipertens&#227;o arterial sist&#234;mica</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>2006</year>
          </date>
          <fpage>e24</fpage>
          <lpage>e99</lpage>
          <issue>3</issue>
          <volume>89</volume>
        </element-citation>
      </ref>
      <ref id="B3">
        <mixed-citation>3. Gus I, Harzheim E, Zaslavsky C, Medina C, Gus M. Prevalence, awareness and control of systemic arterial hypertension in the state of Rio Grande do Sul. Arq Bras Cardiol. 2004;83(5):429-33.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Prevalence, awareness and control of systemic arterial hypertension in the state of Rio Grande do Sul</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>2004</year>
          </date>
          <fpage>429</fpage>
          <lpage>33</lpage>
          <issue>5</issue>
          <volume>83</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Gus</surname>
              <given-names>I</given-names>
            </name>
            <name>
              <surname>Harzheim</surname>
              <given-names>E</given-names>
            </name>
            <name>
              <surname>Zaslavsky</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Medina</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Gus</surname>
              <given-names>M</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B4">
        <mixed-citation>4. Jardim PCBV, Moreira H, Peixoto M, Monego E, Miranda A, Costa C, et al. Knowledge of hypertension and its control in the adult population of a Brazilian state. Am J Hypertens. 2003;16:210A.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Knowledge of hypertension and its control in the adult population of a Brazilian state</article-title>
          <source>Am J Hypertens</source>
          <date>
            <year>2003</year>
          </date>
          <fpage>210A</fpage>
          <volume>16</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Jardim</surname>
              <given-names>PCBV</given-names>
            </name>
            <name>
              <surname>Moreira</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Peixoto</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Monego</surname>
              <given-names>E</given-names>
            </name>
            <name>
              <surname>Miranda</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Costa</surname>
              <given-names>C</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B5">
        <mixed-citation>5. Sociedade Brasileira de Cardiologia. Sociedade Brasileira de Hipertens&#259;o. Sociedade Brasileira de Nefrologia. VI Diretrizes brasileiras de hipertens&#259;o arterial. Arq Bras Cardiol. 2010;93(3 supl. 2):1-112.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>VI Diretrizes brasileiras de hipertens&#227;o arterial</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>2010</year>
          </date>
          <fpage>1</fpage>
          <lpage>112</lpage>
          <issue>3</issue>
          <volume>93</volume>
        </element-citation>
      </ref>
      <ref id="B6">
        <mixed-citation>6. Littenberg B. A practice guideline revisited: screening for hypertension. Ann Intern Med. 1995;122(12):937-9.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>A practice guideline revisited: screening for hypertension</article-title>
          <source>Ann Intern Med</source>
          <date>
            <year>1995</year>
          </date>
          <fpage>937</fpage>
          <lpage>9</lpage>
          <issue>12</issue>
          <volume>122</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Littenberg</surname>
              <given-names>B</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B7">
        <mixed-citation>7. Wolff T, Miller T. Evidence for the reaffirmation of the U.S. Preventive Services Task Force recommendation on screening for high blood pressure. Ann Intern Med. 2007;147(11):787-91.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Evidence for the reaffirmation of the U: S. Preventive Services Task Force recommendation on screening for high blood pressure</article-title>
          <source>Ann Intern Med</source>
          <date>
            <year>2007</year>
          </date>
          <fpage>787</fpage>
          <lpage>91</lpage>
          <issue>11</issue>
          <volume>147</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Wolff</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Miller</surname>
              <given-names>T</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B8">
        <mixed-citation>8. Gomes MAM, Gomes AMC, Vasconcelos JNG, Silva SM, Magalh&#259;es LBNC. Avalia&#231;&#259;o de m&#233;todo para rastreamento e controle de hipertensos. Revista baiana sa&#250;de p&#250;blica. 2006;30(1):169-78.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Avalia&#231;&#227;o de m&#233;todo para rastreamento e controle de hipertensos</article-title>
          <source>Revista baiana sa&#250;de p&#250;blica</source>
          <date>
            <year>2006</year>
          </date>
          <fpage>169</fpage>
          <lpage>78</lpage>
          <issue>1</issue>
          <volume>30</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Gomes</surname>
              <given-names>MAM</given-names>
            </name>
            <name>
              <surname>Gomes</surname>
              <given-names>AMC</given-names>
            </name>
            <name>
              <surname>Vasconcelos</surname>
              <given-names>JNG</given-names>
            </name>
            <name>
              <surname>Silva</surname>
              <given-names>SM</given-names>
            </name>
            <name>
              <surname>Magalh&#227;es</surname>
              <given-names>LBNC</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B9">
        <mixed-citation>9. Forsvall A, Oscarson M, Magalh&#259;es LB, Palmeira C, Guimar&#259;es AC, Gomes MA, et al. Avalia&#231;&#259;o do Rastre&#244;metro, um novo equipamento para triagem populacional de press&#259;o arterial elevada em pa&#237;ses em desenvolvimento. Arq Bras Cardiol. 2006;87(4):480-6.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Avalia&#231;&#227;o do Rastre&#244;metro, um novo equipamento para triagem populacional de press&#227;o arterial elevada em pa&#237;ses em desenvolvimento</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>2006</year>
          </date>
          <fpage>480</fpage>
          <lpage>6</lpage>
          <issue>4</issue>
          <volume>87</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Forsvall</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Oscarson</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Magalh&#227;es</surname>
              <given-names>LB</given-names>
            </name>
            <name>
              <surname>Palmeira</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Guimar&#227;es</surname>
              <given-names>AC</given-names>
            </name>
            <name>
              <surname>Gomes</surname>
              <given-names>MA</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B10">
        <mixed-citation>10. Minist&#233;rio da Sa&#250;de. Mapas e gr&#225;ficos da cobertura da estrat&#233;gia sa&#250;de da fam&#237;lia - Brasil.&#160; 2010&#160; [Acesso em 2010 fev. 16]. Dispon&#237;vel em: <ext-link ext-link-type="uri" xlink:href="http://dab.saude.gov.br/abnumeros.php#historico">http://dab.saude.gov.br/abnumeros.php#historico</ext-link></mixed-citation>
        <element-citation publication-type="book">
          <source>Mapas e gr&#225;ficos da cobertura da estrat&#233;gia sa&#250;de da fam&#237;lia - Brasil</source>
          <date>
            <year>2010</year>
          </date>
        </element-citation>
      </ref>
      <ref id="B11">
        <mixed-citation>11. Automatic Digital Blood Pressure Devices for Clinical Use and also suitable for home/self assessmen. [Acesso em 2008 jun 16]. Dispon&#237;vel em <ext-link ext-link-type="uri" xlink:href="http://www.bhsoc.org/bp_monitors/automatic.stm">http://www.bhsoc.org/bp_monitors/automatic.stm</ext-link></mixed-citation>
        <element-citation publication-type="book">
          <source>Automatic Digital Blood Pressure Devices for Clinical Use and also suitable for home/self assessmen</source>
        </element-citation>
      </ref>
      <ref id="B12">
        <mixed-citation>12. Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves J, Hill MN, et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Hypertension. 2005;45(1):142-61.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research</article-title>
          <source>Hypertension</source>
          <date>
            <year>2005</year>
          </date>
          <fpage>142</fpage>
          <lpage>61</lpage>
          <issue>1</issue>
          <volume>45</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Pickering</surname>
              <given-names>TG</given-names>
            </name>
            <name>
              <surname>Hall</surname>
              <given-names>JE</given-names>
            </name>
            <name>
              <surname>Appel</surname>
              <given-names>LJ</given-names>
            </name>
            <name>
              <surname>Falkner</surname>
              <given-names>BE</given-names>
            </name>
            <name>
              <surname>Graves</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Hill</surname>
              <given-names>MN</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B13">
        <mixed-citation>13. OBrien E, Asmar R, Beilin L, Imai Y, Mallion JM, Mancia G, et al. European Society of Hypertension recommendations for conventional ambulatory and home blood pressure measurement. J Hypertens. 2003;21(5):821-48.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>European Society of Hypertension recommendations for conventional ambulatory and home blood pressure measurement</article-title>
          <source>J Hypertens</source>
          <date>
            <year>2003</year>
          </date>
          <fpage>821</fpage>
          <lpage>48</lpage>
          <issue>5</issue>
          <volume>21</volume>
        </element-citation>
      </ref>
      <ref id="B14">
        <mixed-citation>14. Arcuri EA. Manguito do esfigmoman&#244;metro e diagn&#243;stico de hipertens&#259;o arterial sist&#281;mica. Arq Bras Cardiol. 1989;52(4):181-3.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Manguito do esfigmoman&#244;metro e diagn&#243;stico de hipertens&#227;o arterial sist&#234;mica</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>1989</year>
          </date>
          <fpage>181</fpage>
          <lpage>3</lpage>
          <issue>4</issue>
          <volume>52</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Arcuri</surname>
              <given-names>EA</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B15">
        <mixed-citation>15. Mancia G, Parati G, Pomidossi G, Grassi G, Casadei R, Zanchetti A. Alerting reaction and rise in blood pressure during measurement by physician and nurse. Hypertension. 1987;9(2):209-15.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Alerting reaction and rise in blood pressure during measurement by physician and nurse</article-title>
          <source>Hypertension</source>
          <date>
            <year>1987</year>
          </date>
          <fpage>209</fpage>
          <lpage>15</lpage>
          <issue>2</issue>
          <volume>9</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Mancia</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Parati</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Pomidossi</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Grassi</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Casadei</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Zanchetti</surname>
              <given-names>A</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B16">
        <mixed-citation>16. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care. 1986;24(1):67-74.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Concurrent and predictive validity of a self-reported measure of medication adherence</article-title>
          <source>Med Care</source>
          <date>
            <year>1986</year>
          </date>
          <fpage>67</fpage>
          <lpage>74</lpage>
          <issue>1</issue>
          <volume>24</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Morisky</surname>
              <given-names>DE</given-names>
            </name>
            <name>
              <surname>Green</surname>
              <given-names>LW</given-names>
            </name>
            <name>
              <surname>Levine</surname>
              <given-names>DM</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B17">
        <mixed-citation>17. Guyatt G, Sacket D, Haynes B. Avaliando teste diagn&#243;sticos.&#160; Epidemiologia cl&#237;nica: como realizar pesquisa cl&#237;nica na pr&#225;tica. Porto Alegre: Artmed; 2006. p. 305-56.</mixed-citation>
        <element-citation publication-type="book">
          <source>Avaliando teste diagn&#243;sticos.Epidemiologia cl&#237;nica: como realizar pesquisa cl&#237;nica na pr&#225;tica</source>
          <date>
            <year>2006</year>
          </date>
          <person-group person-group-type="author">
            <name>
              <surname>Guyatt</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Sacket</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Haynes</surname>
              <given-names>B</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B18">
        <mixed-citation>18. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Standards for Reporting of Diagnostic Accuracy. Clin Chem. 2003;49(1):1-6.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Standards for Reporting of Diagnostic Accuracy</article-title>
          <source>Clin Chem</source>
          <date>
            <year>2003</year>
          </date>
          <fpage>1</fpage>
          <lpage>6</lpage>
          <issue>1</issue>
          <volume>49</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Bossuyt</surname>
              <given-names>PM</given-names>
            </name>
            <name>
              <surname>Reitsma</surname>
              <given-names>JB</given-names>
            </name>
            <name>
              <surname>Bruns</surname>
              <given-names>DE</given-names>
            </name>
            <name>
              <surname>Gatsonis</surname>
              <given-names>CA</given-names>
            </name>
            <name>
              <surname>Glasziou</surname>
              <given-names>PP</given-names>
            </name>
            <name>
              <surname>Irwig</surname>
              <given-names>LM</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B19">
        <mixed-citation>19. Fletcher RH, Fletcher SW, Wagner EH. Diagn&#243;stico: epidemiologia cl&#237;nica: aspectos essenciais. 3&#350;. ed. Porto Alegre: Artes M&#233;dicas; 1996. p. 52-82.</mixed-citation>
        <element-citation publication-type="book">
          <source>Diagn&#243;stico: epidemiologia cl&#237;nica: aspectos essenciais</source>
          <date>
            <year>1996</year>
          </date>
          <person-group person-group-type="author">
            <name>
              <surname>Fletcher</surname>
              <given-names>RH</given-names>
            </name>
            <name>
              <surname>Fletcher</surname>
              <given-names>SW</given-names>
            </name>
            <name>
              <surname>Wagner</surname>
              <given-names>EH</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B20">
        <mixed-citation>20. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. The Seventh Report of the Joint National Commitee on Prevention Evaluation and Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003;289(19):2560-72.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>The Seventh Report of the Joint National Commitee on Prevention Evaluation and Treatment of High Blood Pressure: the JNC 7 report</article-title>
          <source>JAMA</source>
          <date>
            <year>2003</year>
          </date>
          <fpage>2560</fpage>
          <lpage>72</lpage>
          <issue>19</issue>
          <volume>289</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Chobanian</surname>
              <given-names>AV</given-names>
            </name>
            <name>
              <surname>Bakris</surname>
              <given-names>GL</given-names>
            </name>
            <name>
              <surname>Black</surname>
              <given-names>HR</given-names>
            </name>
            <name>
              <surname>Cushman</surname>
              <given-names>WC</given-names>
            </name>
            <name>
              <surname>Green</surname>
              <given-names>LA</given-names>
            </name>
            <name>
              <surname>Izzo Jr</surname>
              <given-names>JL</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B21">
        <mixed-citation>21. Izzo JL Jr, Levy D, Black HR. Clinical Advisory Statement: importance of systolic blood pressure in older Americans. Hypertension. 2000;35(5):1021-4.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Clinical Advisory Statement: importance of systolic blood pressure in older Americans</article-title>
          <source>Hypertension</source>
          <date>
            <year>2000</year>
          </date>
          <fpage>1021</fpage>
          <lpage>4</lpage>
          <issue>5</issue>
          <volume>35</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Izzo Jr</surname>
              <given-names>JL</given-names>
            </name>
            <name>
              <surname>Levy</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Black</surname>
              <given-names>HR</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B22">
        <mixed-citation>22. Mancia G, Backe GD, Dominiczak A, Cifkova R, Fagard R, Germano G, et al. 2007 Guidelines for the management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). Eur Heart J. 2007;28(12):1462-536.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>2007 Guidelines for the management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC)</article-title>
          <source>Eur Heart J</source>
          <date>
            <year>2007</year>
          </date>
          <fpage>1462</fpage>
          <lpage>536</lpage>
          <issue>12</issue>
          <volume>28</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Mancia</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Backe</surname>
              <given-names>GD</given-names>
            </name>
            <name>
              <surname>Dominiczak</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Cifkova</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Fagard</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Germano</surname>
              <given-names>G</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B23">
        <mixed-citation>23. Gottschall CAM. Din&#226;mica cardiovascular: do mi&#243;cito &#341; maratona. Porto Alegre: Atheneu; 2005.</mixed-citation>
        <element-citation publication-type="book">
          <source>Din&#226;mica cardiovascular: do mi&#243;cito &#224; maratona</source>
          <date>
            <year>2005</year>
          </date>
          <person-group person-group-type="author">
            <name>
              <surname>Gottschall</surname>
              <given-names>CAM</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
    </ref-list>
  </back>
  <sub-article article-type="translation" id="TRen" xml:lang="en">
    <front-stub>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Original Articles</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title xml:lang="en">Evaluation of accuracy of rastreometer: a new equipment for hypertension tracking</article-title>
      </title-group>
      <abstract xml:lang="en">
        <p>Background: Hypertension is one of the major cardiovascular risk factors affecting more than one billion people. The low awareness of the disease is one of the greatest difficulties related to hypertension. The population screening for hypertension is widely recognized as an effective strategy to detect new cases, but until now, there is no instrument that has become feasible for mass screening of the population. Objective: Evaluate the accuracy and reproducibility of the Rastreometer, a new instrument for the screening of hypertension for community health agents. Methods: Cross-sectional study with a sample of 436 individuals served by the Family Health Strategy. The evaluation was conducted comparing the results of blood pressure obtained by the oscillometric method and reading of the Rastreometer performed by community health agents. Results: The Rastreometer presented a sensitivity of 81.3% and a specificity of 85.6% when assessed in relation to the values of systolic pressure. The operators&#8217; performance influenced the results of the Rastreometer independently (p = 0.0001). The Rastreometer showed substantial reproducibility (Kendall&#8217;s W 0.71). Conclusion: The Rastreometer, when operated by community health agents, showed good sensitivity, specificity and reproducibility, especially when evaluated with the values of systolic pressure, tending to confirm this new instrument as a method of screening for hypertension in primary health care.</p>
      </abstract>
      <kwd-group xml:lang="en">
        <kwd>Hypertension</kwd>
        <kwd>blood pressure</kwd>
        <kwd>measurement equipment</kwd>
        <kwd>sphygmomanometers</kwd>
        <kwd>risk factors</kwd>
        <kwd>diagnosis</kwd>
      </kwd-group>
    </front-stub>
    <body>
      <p>
        <bold>Evaluation of accuracy of rastreometer: a new equipment for hypertension tracking</bold>
      </p>
      <p>
        <bold>Fernando Bitencourt; Carlos Antonio Mascia Gottschall</bold>
      </p>
      <p> Instituto de Cardiologia do Rio Grande do Sul/ Funda&#231;&#227;o Universit&#225;ria de Cardiologia - IC/FUC, Porto Alegre, RS, Brasil</p>
      <p>Mailing Address</p>
      <p>
        <bold>ABSTRACT</bold>
      </p>
      <p><bold>Background</bold>: Hypertension is one of the major cardiovascular risk factors affecting more than one billion people. The low awareness of the disease is one of the greatest difficulties related to hypertension. The population screening for hypertension is widely recognized as an effective strategy to detect new cases, but until now, there is no instrument that has become feasible for mass screening of the population.</p>
      <p><bold>Objective</bold>: Evaluate the accuracy and reproducibility of the Rastreometer, a new instrument for the screening of hypertension for community health agents.</p>
      <p><bold>Methods</bold>: Cross-sectional study with a sample of 436 individuals served by the Family Health Strategy. The evaluation was conducted comparing the results of blood pressure obtained by the oscillometric method and reading of the Rastreometer performed by community health agents.</p>
      <p><bold>Results</bold>: The Rastreometer presented a sensitivity of 81.3% and a specificity of 85.6% when assessed in relation to the values of systolic pressure. The operators&#146; performance influenced the results of the Rastreometer independently (p = 0.0001). The Rastreometer showed substantial reproducibility (Kendall&#146;s W 0.71).</p>
      <p><bold>Conclusion</bold>: The Rastreometer, when operated by community health agents, showed good sensitivity, specificity and reproducibility, especially when evaluated with the values of systolic pressure, tending to confirm this new instrument as a method of screening for hypertension in primary health care.</p>
      <p><bold>Keywords</bold>: Hypertension / prevention, blood pressure, measurement equipment, sphygmomanometers, risk factors, diagnosis.</p>
      <p>
        <bold>Introduction</bold>
      </p>
      <p>Hypertension (HBP) is one of the major cardiovascular risk factors and major risk factor attributed to global mortality<sup>1</sup>. Early detection is paramount in the treatment of hypertension in order to reduce cardiovascular events<sup>2</sup>. However, the low rates of knowledge of the disease are one of the greatest difficulties in its management<sup>3,4</sup>. In Brazil, it is estimated that half of hypertensive individuals know their condition<sup>5</sup>. The population screening for hypertension is recognized as an effective strategy in the detection of new cases<sup>6,7</sup>.</p>
      <p>So far, however, does there is no instrument available that makes it feasible for mass screening of the population.</p>
      <p>Studies<sup>8,9</sup> have shown that a new instrument, the Rastreometer could be a possibility for population screening for hypertension in primary care, since it was designed to be used by Community Health Agents (CHA). This means that more than 240,000 ACS<sup>10</sup> across the country could apply this test quickly, with low costs and minimal risks, to detect suspected hypertension individuals. However, the Rastreometer has only been tested by professionals and medical students in controlled environments<sup>8,9</sup>.</p>
      <p>The Rastreometer (<xref ref-type="fig" rid="ea08fig01_-body2">Fig. 1</xref>) is an adaptation of an aneroid sphygmomanometer in which the gauge display is covered by a sticker with two bands, one yellow indicating pressure values &#8203;&#8203;&lt;140 mmHg and another red indicating values &#8203;&#8203;= or&gt; 140 mmHg. The evaluation is done by reading the beginning of the oscillation of the needle gauge, being suspected of HBP that in which the oscillation of the needle start in the red band, and those with normal pressure in which oscillation starts in the yellow band. A limitation of this technique is the impossibility, <italic>a priori</italic>, to evaluate isolated changes in diastolic pressure.</p>
      <p content-type="asset">
        <fig id="ea08fig01_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gf01-en.jpg"/></fig>
      </p>
      <p>This technique was developed because of the prohibition, by the Federal Nursing Council, of the CHA perform measurements of blood pressure by auscultation / oscillometric method.</p>
      <p>The objective of this study is to evaluate the accuracy and reproducibility of the Rastreometer as a screening method for HBP when operated by community health agents.</p>
      <p>
        <bold>Methods</bold>
      </p>
      <p>Cross-sectional study in the town of Charqueadas (RS). The study population was the inhabitants ascribed in the <italic>Unidade de Sa&#250;de da Fam&#237;lia Sul Am&#233;rica.</italic> Survey participants were selected by simple random probability sampling.</p>
      <p>The sample size calculation to evaluate the accuracy of the Rastreometer was conducted based on the values &#8203;&#8203;of sensitivity (95%) and specificity (63.1%) presented in previous studies and on the prevalence of hypertension in the study area (21%). We allowed a margin of error of 5% and 80% power. The result of the sample calculation was 436 people.</p>
      <p>In order to test the reproducibility of the Rastreometer, a second sample of 90 patients was evaluated. For this test, we used a convenience sample of people who sought treatment in a second Health Unit of the municipality.</p>
      <p>Inclusion criteria were: arm circumference between 22 cm and 42 cm, regular use of antihypertensive drug therapy in patients with hypertension and aged between 20 and 80 years.</p>
      <p>The Rastreometer method was compared with the oscillometric method, considered a reference standard in the study. The Rastreometer was made from an aneroid sphygmomanometer brand <italic>Eternum</italic> adult size (rubber bag 12 cm wide and 23 cm in length), indicated for brachial circumferences between 24 and 30 cm. As reference standard it was used a digital oscillometric sphygmomanometer brand <italic>Microlife</italic> 3BTO-A model, validated by the British Hypertension Society<sup>11</sup>. In this device, we used two types of clamps with appropriate sizes for each arm circumference (22-32cm Standard adult and Large adult 32-42cm). This digital apparatus was used as a comparison in an attempt to avoid measurement biases<sup>12,13</sup>.</p>
      <p>Prior to beginning work, the tools of research, which were new and certified by the National Institute of Metrology, Standardization and Industrial Quality - <italic>Inmetro</italic> - (with certification seal of May 2008), were evaluated to ascertain suitability of the calibration between sphygmomanometers. After the beginning of the survey we conducted other two calibration checks. The calibration evaluations were performed by Inmetro. In none of the assessments there was a discrepancy regarding calibration of the instruments or disapproval of any other item evaluated.</p>
      <p>The Rastreometer was used in participants with arm circumference between 22 cm and 42 cm. To evaluate the blood pressure it was used a specific clamp for the different brachial circumferences, following the manufacturer&#146;s specifications of the Microlife sphygmomanometer. This option for comparing a blood pressure level closest to the actual value (by the use of appropriate clamps for arm circumference) with that obtained by reading the Rastreometer (only with adult-type device) was used to evaluate the influence of arm circumference in the accuracy of the Rastreometer, since the size of the cuff has unquestionable influence on blood pressure measurements by the indirect methods<sup>14</sup>.</p>
      <p>The Rastreometer method was handled by four CHA and their readings were compared with the oscillometric method, whose measurements were taken using a nurse<sup>15</sup>.</p>
      <p>All CHA used the same equipment in all phases of research. The CHA were not familiar with the new method, therefore, they received a basic training in the technique of reading the Rastreometer and later each one performed 30 measurements with the instrument in different patients.</p>
      <p>Initially, we applied the questionnaires and obtained anthropometric measurements for each participant. The interview, anthropometric measurements and evaluation of blood pressure by the oscillometric method occurred only in the presence of the nurse in order to keep the CHA blind to the clinical characteristics and values &#8203;&#8203;of blood pressure. The operators of the Rastreometer and reference standard device were also blind to each other.</p>
      <p>Each participant had their blood pressure measured twice by the oscillometric method and twice with the Rastreometer. Firstly, evaluations were made by the oscillometric method, and after, with the Rastreometer. All assessments of blood pressure followed the orientation of the V Brazilian Guidelines on Arterial Hypertension<sup>2</sup>. The evaluations were performed after an initial rest for 5 minutes and 1 minute between the first and second measurements (reference standard method) and between the third and fourth measurements (Rastreometer). Between the second and third measures there was an interval of 2 minutes<sup>2</sup>.</p>
      <p>It was considered positive for high blood pressure using the Rastreometer if one or both of the readings taken were positive (oscillation of the needle in the red band indicating systolic blood pressure equal to or greater than 140 mmHg). To compare the results, it was considered the arithmetic mean of blood pressure values &#8203;&#8203;obtained with the oscillometric method. We considered high blood pressure values &#8203;&#8203;greater than or equal to 140 mmHg for systolic pressure and 90 mmHg for the diastolic pressure<sup>2</sup>. The regular use of drug therapy was evaluated by the Morisky-Green<sup>16 </sup>questionnaire.</p>
      <p>To evaluate the reproducibility of the Rastreometer it was carried out a new sequence of evaluations in an independent sample. In this test, each CHA held a Rastreometer measured with an interval of one minute between each measure in the same participant. In this evaluation it was also obtained a blood pressure measurement by the oscillometric method. During the reproducibility tests, the CHA were blinded to each other, for the prior diagnosis of hypertension and blood pressure value.</p>
      <p>We used Student&#146;s t test to compare continuous variables and chi-square test for categorical variables. To identify the variables that independently influenced the results of the Rastreometer, we performed bivariate and multivariate analysis. Initially, all variables of interest were tested in bivariate model, as follows: age, gender, race, body mass index (BMI&gt; 30 / BMI &lt;30), heart rate, cuff size used in Rastreometer (suitable / unsuitable to arm circumference), previous diagnosis of hypertension, diabetes or dyslipidemia, smoking and operator (comparison of the percentage of success among operators).</p>
      <p>For accuracy assessments we performed analysis for sensitivity, specificity, positive and negative predictive value, pre-test probability, likelihood ratio +, likelihood ratio -, post-test probability+ and post-test probability<sup>17</sup>. To evaluate intra-observer concordance, it was used the Kappa test, and to assess the concordance between the observers it was used the Kendall&#146;s W test. It was considered as statistical significance a p value &lt;0.05 and it was used a range of 95%. The power of statistical tests was 80%.</p>
      <p>The protocol was approved by the Research Ethics Committee and all participants signed an informed consent.</p>
      <p>
        <bold>Results</bold>
      </p>
      <p>The first visit took place on 08/23/2008 and the last on 7/24/2009. The study flowchart is shown in Figure 2<sup>18</sup>.</p>
      <p>The sample was predominantly female (61.9%) and white (77.8%). The average age was 47 years. The total number of patients with a previous diagnosis of hypertension was 131, with the percentage of hypertensive patients with controlled pressure of 58%. Eighty-one patients had high blood pressure (BP) during the test. The other clinical and demographic characteristics of the sample are presented in <xref ref-type="table" rid="ea08tab01_-body2">Table 1</xref>.</p>
      <p content-type="asset">
        <table-wrap id="ea08tab01_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt01-en.jpg"/></table-wrap>
      </p>
      <p>Once the Rastreometer has the ability, <italic>a priori</italic>, to identify changes in the systolic component of blood pressure, allowing only the tracking of cases of isolated systolic hypertension and systolic / diastolic pressure, the results described below will be presented, first considering hypertension as the reference values &#8203;&#8203;of the VI Brazilian Guidelines on Arterial Hypertension<sup>5</sup>, and it will be described as assessment of systolic and diastolic pressure (ASDP) and second, considering only the value of systolic pressure equal to or greater than 140 mmHg, characterizing hypertension, discarding cases of isolated diastolic hypertension, and it will be described as Assessment of Systolic Pressure (ASP).</p>
      <p>The results for ASDP and ASP are presented in <xref ref-type="table" rid="ea08tab02_-body2">Table 2</xref>. By evaluating the values &#8203;&#8203;of isolated systolic pressure there was an increase of 10.8% in the value of sensitivity and slight reduction in specificity (1.7%). Regarding predictive values, there was a reduction in positive predictive value (8.3%), as expected due to the reduction of prevalence, and a slight increase in the negative predictive value (3.9%). However, the likelihood for positive and negative ratio was only marginally altered, with a slight increase in ASP.</p>
      <p content-type="asset">
        <table-wrap id="ea08tab02_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt02-en.jpg"/></table-wrap>
      </p>
      <p>The performance of each CHA was also evaluated considering ASP and ASDP. The results are shown in <xref ref-type="table" rid="ea08tab03_-body2">Tables 3</xref> and <xref ref-type="table" rid="ea08tab04_-body2">4</xref>. We noted that there was an improvement in the results of sensitivity and slight reduction in specificity also in the operators&#146; individual results when considering the ASP. The differences between the agents, however, persisted in both evaluations.</p>
      <p content-type="asset">
        <table-wrap id="ea08tab03_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt03-en.jpg"/></table-wrap>
      </p>
      <p content-type="asset">
        <table-wrap id="ea08tab04_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt04-en.jpg"/></table-wrap>
      </p>
      <p>The number of cases of isolated diastolic hypertension (22 patients had isolated elevation of diastolic pressure) and the interpretation of each CHA are presented in <xref ref-type="table" rid="t05-body2">Table 5</xref>.</p>
      <table-wrap id="t05-body2">
        <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt05-en.jpg"/></table-wrap>
      <p>The reproducibility of the Rastreometer was assessed by interobserver agreement by Kendall&#146;s W test with result of 0.71. Measurements of intra-observer concordance were also performed for each CHA through the Kappa test (<xref ref-type="table" rid="a08tab03_-body2">Tables 3</xref> and <xref ref-type="table" rid="ea08tab04_-body2">4</xref>). A general measure of concordance was performed to evaluate the overall Rastreometer concordance with the reference standard method of obtaining a Kappa value of 0.52 (CI 95% 0.43 to 0.62) for ASDP and Kappa of 0.51 (CI 95% 0.40 to 0.62) for ASP.</p>
      <p content-type="asset">
        <table-wrap id="a08tab03_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gt06-en.jpg"/></table-wrap>
      </p>
      <p>Also learning curves were generated for health agents in order to evaluate the performance of each operator throughout the study. Comparisons were made between the percentages of success between quintiles of samples from each CHA. These showed that none of the operators varied over the course of their performance evaluations (p = ns).</p>
      <p>In the Poisson regression analysis, the only variable that independently influenced the results of the Rastreometer was the operator (p = 0.0001). The use of the sphygmomanometer cuff inadequate for arm circumference (p = 0.19) and previous diagnosis of hypertension (p = 0.10) did not influence the results of Rastreometer significantly.</p>
      <p>
        <bold>Discussion</bold>
      </p>
      <p>This was the first work with the Rastreometer involving community health agents to assess its accuracy as a screening test for hypertension that we know. In this study, the Rastreometer showed good sensitivity and specificity, especially when weighed against the values &#8203;&#8203;of systolic pressure. The use of inappropriate cuff for arm circumference in the Rastreometer (p = 0.19) and previous diagnosis of hypertension (p = 0.10) did not influence the results of the new instrument. The performance of operators (CHA) influenced the results of the Rastreometer independently (p = 0.0001). The Rastreometer showed substantial reproducibility as measured by Kendall&#146;s W test (0.71).</p>
      <p>This was the third study involving this new method for screening of hypertension; however, it was the first to evaluate the accuracy of the Rastreometer with CHA and in a primary care environment.</p>
      <p>In this evaluation the Rastreometer had a sensitivity of 70.5% (CI 95% 57.0 to 79.7) and specificity of 87.3% (CI 95% 83.3 to 90.5) when evaluated regarding systolic and diastolic pressure values (ASDP). Considering only the values &#8203;&#8203;of systolic blood pressure (ASP) the Rastreometer increased the sensitivity to 81.3% (CI 95% 68.6 to 89.8) and had a small reduction in its specificity which has been established in 85.6% (CI 95% from 81.6 to 88.9).</p>
      <p>Our findings corroborate the results of previous studies. In the first study<sup>8</sup> the Rastreometer was tested by its own inventor in a sample of 200 patients. Of these, 100 were known to be without a diagnosis of hypertension and 100 known to be hypertensive. The results of sensitivity and specificity for the group without a diagnosis of hypertension were quite high, reaching values &#8203;&#8203;of 96.7% to 80% for sensitivity and specificity. On the other hand, in the group diagnosed with the disease, the sensitivity was 98.7% and a specificity of 94.7%. In the second study<sup>9</sup> the Rastreometer was tested on a sample of 268 patients, by two medical students, with results of 95.1% for sensitivity and 63.1% for specificity.</p>
      <p>Although the results of the three studies tend to confirm the Rastreometer as a method of population screening for hypertension, some considerations are relevant.</p>
      <p>Both previous studies were performed in secondary care environments. Thus, the results may have been overestimated by the characteristics of the populations studied, namely, higher prevalence of hypertension and the spectrum of illness of patients in these studies<sup>17</sup>. By themselves, these two factors could increase the sensitivity, due to the spectrum bias, and positive predictive value, related to higher prevalence of disease<sup>19</sup>.</p>
      <p>The lack of blinding of the Rastreometer operators to the clinical characteristics of the participants in previous studies could also be related to the increased sensitivity of the instrument under test, since the knowledge of the patient&#146;s diagnosis before applying the test may take, even if unconsciously, to identify patients as patients with a previously established diagnosis<sup>17,19</sup>. Another important fact was the number of patients in which the Rastreometer was applied in previous studies, which could result in random bias<sup>19</sup>.</p>
      <p>It is worth noting that the characteristics of the operators of the three studies are fairly distinct. The Rastreometer was operated by an experienced cardiologist and inventor of the instrument in the first study, by two medical students in the second study and, finally, four CHA with high school degree. This fact could also result in differences between the results of studies supporting the aforementioned<sup>17</sup>.</p>
      <p>Another important factor was the diagnostic accuracy used in the different studies. In the study of Forsvall<sup>9</sup> it was used reference standard with an accuracy of 5 mmHg. In our study, the accuracy in measurements of blood pressure was 1 mmHg. The accuracy of BP measurements in the first study was not reported.</p>
      <p>Thus, the performance of the Rastreometer in this study is consistent with that described by Fletcher<sup>19</sup>, which reinforces that being applied in an asymptomatic population and in milder cases of disease, the screening tests tend to have lower sensitivity and higher specificity than when applied to a spectrum of more severe patients, and the positive predictive value is generally low, even with high values &#8203;&#8203;of specificity.</p>
      <p>The assessment carried out in relation to isolated systolic BP values &#8203;&#8203;is quite relevant, since the Rastreometer has an intrinsic limitation, which is the ability to evaluate only the systolic pressure gauge (start of oscillation of the needle). Systolic hypertension, however, is the most prevalent form of hypertension and the most important cardiovascular risk factor<sup>20</sup>. Nevertheless, the isolated value of the systolic blood pressure component is able to correctly identify 91% of arterial hypertension cases<sup>21</sup>.</p>
      <p>In this study, 24 patients with abnormal blood pressure were not identified by the Rastreometer (<xref ref-type="fig" rid="ea08fig02_-body2">Fig. 2</xref>), and out of these, 11 had elevated systolic BP and 13 had isolated elevation of diastolic BP. However, the Rastreometer identified 9 patients with isolated change in diastolic BP, of a total of 22 cases, which was not expected to occur (<xref ref-type="table" rid="t05-body2">Table 5</xref>). At first glance, the non identification of 24 suspects of hypertension could be considered significant; however, considering the context in which it is intended to use the instrument, the strategy of family health, we will understand that in cases not screened the first time may be identified following the screening, since the test can be applied monthly by community health agents.</p>
      <p content-type="asset">
        <fig id="ea08fig02_-body2">
          <graphic xlink:href="1678-4170-abc-98-02-0151-00013-gf02-en.jpg"/></fig>
      </p>
      <p>In turn, 45 patients (<xref ref-type="fig" rid="ea08fig02_-body2">Fig. 2</xref>) were identified as false-positives by the Rastreometer. It is noteworthy that the identification of false-positive cases would have a benign result; which would be the individuals going to the Health Unit for assessment of blood pressure and, if necessary, clarifying diagnosis in medical consultation.</p>
      <p>The use of a single cuff size, regardless of arm circumference, independently did not affect the properties of the Rastreometer (p = 0.19). Reduced accuracy of the instrument in situations of inadequate cuff was expected, since the influence of cuff size on blood pressure measurement is well defined<sup>12,14,22</sup>. However, this situation can be explained in part by the fact of having been used a cuff smaller than indicated in the Rastreometer in 80.7% of the sample, which increased its sensitivity to some extent due to an overestimation of the arterial pressure<sup>14 </sup>reading and did not influence the specificity. This finding confirms the finding of Forsvall<sup>9</sup>, which also has not identified differences in accuracy of the Rastreometer when using cuffs inadequate to arm circumference, however, in that study we used identical cuffs in the Rastreometer and a reference standard.</p>
      <p>It is worth noting that the evaluation of the influence of arm circumference in the Rastreometer accuracy is impossible to achieve if using cuffs inadequate for arm circumference in the reference standard instrument, as the result obtained in this unit would not be consistent with the reality, but compatible with the Rastreometer measures. The fact that a possible overestimation of values in the Rastreometer sensitivity, by using a cuff smaller that required in the Rastreometer by far is negative, once it is expected high sensitivity values &#8203;&#8203;of screening tests.</p>
      <p>In our study, the previous diagnosis of hypertension did not influence significantly either the properties of the Rastreometer (p = 0.10). However, in the work of Forsvall<sup>9</sup>, when comparing the readings altered in the Rastreometer with the results of blood pressure above 140 mmHg systolic or 90 mmHg diastolic BP, the use of medication for treatment of hypertension, which implies being hypertensive, altered the Rastreometer specificity of 77.8% in the unmedicated group to 32.7% in the medicated group. The regular use of antihypertensive medication was assessed in our study with the aim of identifying variations in drug-related accuracy by comparing hypertensive patients under pharmacological and non-pharmacological isolated treatment. However, there were no cases of non-pharmacological treatment isolated in the sample, making it impossible to evaluate the influence of the use of antihypertensive drugs.</p>
      <p>One of the possibilities to explain this difference between studies was the lack of blinding in the diagnosis of the patient before the measurements with the Rastreometer in that study, which may have led to a tendency to recognize the test applied to patients known to be hypertensive as altered, even when they had controlled pressure, thus incurring the observed reduction in the values of specificity.</p>
      <p>The explanation of the author to justify that difference between the results of specificity in the medicated and unmedicated groups was attributed to the lack of standardization of the beginning of the oscillations that identify the actual systolic pressure, and a better training of operators could correct this distortion. In that study, the operators of the Rastreometer reached their plateau of performance only at the end of the study. This fact is not supported by our findings, since the CHA did not change significantly their performance during data collection, showing that the agents reached a plateau of learning even in the training phase.</p>
      <p>In turn, the operator variable was identified as an independent factor in the variation of results of the Rastreometer (p = 0.0001), showing that the CHA varied their performance amongst themselves. This marked difference between one CHA and another regarding the performance with the Rastreometer indicates that results of the instrument may be linked to the skills of each operator, since there was no difference between the blood pressure means amongst the samples of each CHA (p&gt; 0.05).</p>
      <p>Another hypothesis for the large variability between operators was that the gauge needle does not maintain a definite pattern of oscillation in all patients, making it difficult to pinpoint the exact moment when the blood flow starts to flow through the artery and move the gauge needle, which would be the marker of systolic pressure in the Rastreometer. This lack of a definite pattern of oscillation can be attributed to variations in the patient&#146;s own endogenous characteristics such as volume of blood ejected into the aorta, myocardial contractility and aortic rigidity, since these three factors are determinants of systolic blood pressure<sup>23</sup>. The lack of a definite pattern of oscillation had already been commented upon by Forsvall<sup>9</sup>. The small number of cases of high blood pressure in each sample may have emphasized the variability between the CHA by the random bias<sup>19</sup>.</p>
      <p>Although there was variation in performance between the CHA when evaluated in independent samples, the test conducted on a second sample to determine the reproducibility of the method presented the result of substantial concordance (Kendall&#146;s W 0.71). The fact that there is less disagreement among operators when evaluating the same patients in a convenience sample reinforces the influence of the disease spectrum in the accuracy of the instrument, since there was no difference in the mean blood pressure between the accuracy of testing samples (122.7 &#177; 17.5) and reproducibility (126.9 &#177; 20.6) with p-value = 0.04.</p>
      <p>Evaluations of blood pressure by the methods reference standard and rastreometer performed in non-random sequence was a potential source of bias in our study, by the possibility of overestimation of the first measurements (reference standard), which would result in reduced sensitivity. However, the assessment of arterial BP made by the nurse may have minimized the white coat effect<sup>15</sup>, since there was no difference between the means of the second and the first measurements of blood pressure (p&gt; 0.05).</p>
      <p>
        <bold>Conclusion</bold>
      </p>
      <p>The Rastreometer, when operated by community health agents, showed good sensitivity, specificity and reproducibility, especially when evaluated the values &#8203;&#8203;of systolic pressure, tending to confirm this new instrument as a method of screening for hypertension in primary health care. The Rastreometer could be used as a tool in the daily work of community health agents assisting in early identification of hypertensive patients, thus, enabling an improvement in the rates of the disease, today one of the greatest difficulties related to hypertension.</p>
      <p>Other studies involving health agents in other contexts are needed to validate this new tool. The evaluation of the effectiveness of Rastreometer usage should also be investigated by future randomized studies.</p>
      <p>
        <bold>References</bold>
      </p>
    </body>
  </sub-article>
</article>
