<!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">zwb6ZdCrX9ZSDLBnb4fL58C</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v2">S0066-782X2013003000006</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v1">S0066-782X(13)10100400006</article-id>
      <article-id pub-id-type="doi">10.5935/abc.20130181</article-id>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Artigos Originais</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>M&#233;todos n&#227;o-lineares para avaliar mudan&#231;as na variabilidade da frequ&#234;ncia card&#237;aca em pacientes com diabetes tipo 2</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <name>
            <surname>Roy</surname>
            <given-names>Bhaskar</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
          <xref ref-type="aff" rid="aff02"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Ghatak</surname>
            <given-names>Sobhendu</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
        </contrib>
      </contrib-group>
      <aff id="aff01">
        <addr-line>
          <named-content content-type="state"/>
        </addr-line>
        <institution content-type="orgname">Indian Institute of Technology</institution>
        <country country="IN">India</country>
      </aff>
      <aff id="aff02">
        <addr-line>
          <named-content content-type="city">Farmington</named-content>
          <named-content content-type="state"/>
        </addr-line>
        <institution content-type="orgname">University of Connecticut</institution>
        <country country="US">United States</country>
      </aff>
      <pub-date date-type="collection" publication-format="electronic">
        <month>10</month>
        <year>2013</year>
      </pub-date>
      <volume>101</volume>
      <issue>4</issue>
      <fpage>317</fpage>
      <lpage>327</lpage>
      <history>
        <date date-type="received">
          <day>17</day>
          <month>07</month>
          <year>2012</year>
        </date>
        <date date-type="accepted">
          <day>26</day>
          <month>04</month>
          <year>2013</year>
        </date>
        <date date-type="rev-recd">
          <day>09</day>
          <month>09</month>
          <year>2012</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 variabilidade da frequ&#234;ncia card&#237;aca (VFC) &#233; um importante indicador da modula&#231;&#227;o auton&#244;mica da fun&#231;&#227;o cardiovascular. A diabetes pode alterar a modula&#231;&#227;o auton&#244;mica danificando as entradas aferentes, dessa forma aumentando o risco de doen&#231;as cardiovasculares. Foram aplicados m&#233;todos anal&#237;ticos n&#227;o lineares para identificar os par&#226;metros associados com VFC indicativos de altera&#231;&#245;es na modula&#231;&#227;o auton&#244;mica da fun&#231;&#227;o card&#237;aca em pacientes diab&#233;ticos. OBJETIVO: Analisamos as diferen&#231;as nos padr&#245;es da VFC entre pacientes diab&#233;ticos e controles saud&#225;veis pareados por idade, utilizando m&#233;todos n&#227;o-lineares. M&#201;TODOS: Plot de Poincar&#233; Lagged, autocorrela&#231;&#227;o e an&#225;lise de flutua&#231;&#227;o destendenciada foram aplicados para analisar a VFC em registros de eletrocardiograma (ECG). RESULTADOS: A an&#225;lise do gr&#225;fico de Poincar&#233; lagged revelou altera&#231;&#245;es significativas em alguns par&#226;metros, sugestivas de diminui&#231;&#227;o da modula&#231;&#227;o parassimp&#225;tica. O expoente de flutua&#231;&#227;o destendencionada derivado de um ajuste em longo prazo foi maior que o expoente em curto prazo na popula&#231;&#227;o diab&#233;tica, o que tamb&#233;m foi consistente com a diminui&#231;&#227;o do input parassimp&#225;tico. A fun&#231;&#227;o de autocorrela&#231;&#227;o do desvio dos intervalos inter-batimento exibiu um padr&#227;o altamente correlacionado no grupo de diab&#233;ticos em compara&#231;&#227;o com o grupo controle. CONCLUS&#195;O: O padr&#227;o de VFC difere significativamente entre pacientes diab&#233;ticos e indiv&#237;duos saud&#225;veis. Os tr&#234;s m&#233;todos estat&#237;sticos utilizados no estudo podem ser &#250;teis para detectar o in&#237;cio e a extens&#227;o da neuropatia auton&#244;mica em pacientes diab&#233;ticos.</p>
      </abstract>
      <kwd-group xml:lang="pt" kwd-group-type="author-generated">
        <kwd>Frequ&#234;ncia Card&#237;aca</kwd>
        <kwd>Diabetes Mellitus Tipo 1</kwd>
        <kwd>S&#237;stole</kwd>
        <kwd>Medidas, M&#233;todos e Teorias</kwd>
        <kwd>Estat&#237;stica como Assunto</kwd>
      </kwd-group>
      <pub-date publication-format="electronic" date-type="pub">
        <day>06</day>
        <month>09</month>
        <year>2013</year>
      </pub-date>
    </article-meta>
  </front>
  <body>
    <p>
      <bold>M&#233;todos n&#227;o-lineares para avaliar mudan&#231;as na variabilidade da frequ&#234;ncia card&#237;aca em pacientes com diabetes tipo 2</bold>
    </p>
    <p>
      <bold>Bhaskar Roy<sup>I,II</sup>; Sobhendu Ghatak<sup>I</sup></bold>
    </p>
    <p><sup>I</sup>Indian Institute of Technology, &#205;ndia </p>
    <p><sup>II</sup>University of Connecticut, Farmington, CT, USA</p>
    <p>
      <xref ref-type="corresp" rid="end">Correspond&#234;ncia</xref>
    </p>
    <p>
      <bold>RESUMO</bold>
    </p>
    <p><bold>FUNDAMENTO:</bold> A variabilidade da frequ&#234;ncia card&#237;aca (VFC) &#233; um importante indicador da modula&#231;&#227;o auton&#244;mica da fun&#231;&#227;o cardiovascular. A diabetes pode alterar a modula&#231;&#227;o auton&#244;mica danificando as entradas aferentes, dessa forma aumentando o risco de doen&#231;as cardiovasculares. Foram aplicados m&#233;todos anal&#237;ticos n&#227;o lineares para identificar os par&#226;metros associados com VFC indicativos de altera&#231;&#245;es na modula&#231;&#227;o auton&#244;mica da fun&#231;&#227;o card&#237;aca em pacientes diab&#233;ticos. </p>
    <p><bold>OBJETIVO:</bold> Analisamos as diferen&#231;as nos padr&#245;es da VFC entre pacientes diab&#233;ticos e controles saud&#225;veis pareados por idade, utilizando m&#233;todos n&#227;o-lineares. </p>
    <p><bold>M&#201;TODOS:</bold> Plot de Poincar&#233; Lagged, autocorrela&#231;&#227;o e an&#225;lise de flutua&#231;&#227;o destendenciada foram aplicados para analisar a VFC em registros de eletrocardiograma (ECG). </p>
    <p><bold>RESULTADOS:</bold> A an&#225;lise do gr&#225;fico de Poincar&#233; lagged revelou altera&#231;&#245;es significativas em alguns par&#226;metros, sugestivas de diminui&#231;&#227;o da modula&#231;&#227;o parassimp&#225;tica. O expoente de flutua&#231;&#227;o destendencionada derivado de um ajuste em longo prazo foi maior que o expoente em curto prazo na popula&#231;&#227;o diab&#233;tica, o que tamb&#233;m foi consistente com a diminui&#231;&#227;o do <italic>input</italic> parassimp&#225;tico. A fun&#231;&#227;o de autocorrela&#231;&#227;o do desvio dos intervalos inter-batimento exibiu um padr&#227;o altamente correlacionado no grupo de diab&#233;ticos em compara&#231;&#227;o com o grupo controle. </p>
    <p><bold>CONCLUS&#195;O:</bold> O padr&#227;o de VFC difere significativamente entre pacientes diab&#233;ticos e indiv&#237;duos saud&#225;veis. Os tr&#234;s m&#233;todos estat&#237;sticos utilizados no estudo podem ser &#250;teis para detectar o in&#237;cio e a extens&#227;o da neuropatia auton&#244;mica em pacientes diab&#233;ticos.</p>
    <p><bold>Palavras-chave:</bold> Frequ&#234;ncia Card&#237;aca; Diabetes Mellitus Tipo 1; S&#237;stole; Medidas, M&#233;todos e Teorias; Estat&#237;stica como Assunto.</p>
    <p>
      <bold>Introdu&#231;&#227;o</bold>
    </p>
    <p>O ritmo card&#237;aco &#233; dinamicamente regulado por sistemas de controle intr&#237;nsecos e extr&#237;nsecos, mantendo a homeostase. O principal controle extr&#237;nseco &#233; fornecido pelo sistema nervoso aut&#244;nomo. A variabilidade da frequ&#234;ncia card&#237;aca (VFC) &#233; uma medida da flutua&#231;&#227;o no intervalo entre batimentos card&#237;acos sequenciais no ritmo sinusal, e reflete a regula&#231;&#227;o auton&#244;mica card&#237;aca<sup>1-3</sup>. A diabetes leva &#224; neuropatia auton&#244;mica<sup>4</sup>, interrompendo, assim, um importante componente da regula&#231;&#227;o cardiovascular e contribuindo para um aumento da incid&#234;ncia de doen&#231;as cardiovasculares em pacientes diab&#233;ticos, tais como ataque card&#237;aco, morte s&#250;bita card&#237;aca, e isquemia silenciosa<sup>5-8</sup>. O diagn&#243;stico precoce de neuropatia auton&#244;mica diab&#233;tica &#233; dif&#237;cil e os m&#233;todos de detec&#231;&#227;o dispon&#237;veis, por exemplo, os Testes de Ewing, s&#227;o complicados e t&#234;m pouca sensibilidade e reprodutibilidade. Por outro lado, a an&#225;lise da VFC &#233; n&#227;o-invasiva e os dados de entrada s&#227;o facilmente obtidos atrav&#233;s de eletrocardiograma (ECG)<sup>9-12</sup>. No entanto, por causa da din&#226;mica n&#227;o-linear do cora&#231;&#227;o, o tempo convencional e os par&#226;metros de dom&#237;nio da frequ&#234;ncia da VFC podem n&#227;o representar sempre as caracter&#237;sticas n&#227;o-estacion&#225;rias do ECG. M&#233;todos n&#227;o-lineares, como o <italic>plot</italic> de Poincar&#233;, a an&#225;lise de flutua&#231;&#227;o destendenciada (AFD), an&#225;lise de t&#244;nus/entropia, e a an&#225;lise de complexidade da FC s&#227;o ferramentas rec&#233;m-desenvolvidas para identificar padr&#245;es n&#227;o-lineares dos dados do ECG<sup>13-18</sup>.</p>
    <p>Nesse estudo, foram utilizados m&#233;todos anal&#237;ticos n&#227;o-lineares para avaliar as diferen&#231;as nos padr&#245;es da VFC entre indiv&#237;duos diab&#233;ticos e saud&#225;veis. O objetivo deste estudo foi identificar novos par&#226;metros &#250;teis para a detec&#231;&#227;o da desregula&#231;&#227;o auton&#244;mica em diabetes.</p>
    <p>
      <bold>M&#233;todos</bold>
    </p>
    <p>O grupo de pacientes consistiu em 23 pacientes com diabetes mellitus tipo 2 sem hist&#243;rico de doen&#231;as card&#237;acas, neurol&#243;gicas, psiqui&#225;tricas ou dist&#250;rbios de sono. Pacientes recebendo medicamentos que alteram o batimento card&#237;aco foram exclu&#237;dos do estudo. O estudo foi aprovado pelo comit&#234; de &#233;tica do <italic>Indian Institution of Technology</italic> (Instituto Indiano de Tecnologia), Kharagpur, &#205;ndia. Um total de 23 indiv&#237;duos saud&#225;veis foi selecionado como grupo controle, utilizando os mesmos crit&#233;rios de exclus&#227;o. Todos os participantes deram consentimento informado por escrito antes da inclus&#227;o no estudo. Os participantes foram instru&#237;dos a evitar cafe&#237;na, &#225;lcool e exerc&#237;cios f&#237;sicos no dia anterior ao estudo. Um registro de 10 min do ECG foi feito nos pacientes em posi&#231;&#227;o supina ap&#243;s um per&#237;odo de descanso de 15 min. Todos os eletrocardiogramas foram registrados em um per&#237;odo fixo do dia para evitar os efeitos das varia&#231;&#245;es diurnas na VFC.</p>
    <p>Foram utilizados para a an&#225;lise estat&#237;stica os pacotes de <italic>software</italic> Matlab e SPSS. Para a an&#225;lise comparativa entre os grupos, os testes <italic>t</italic> n&#227;o pareados foram aplicados quando apropriados. Outros m&#233;todos estat&#237;sticos s&#227;o descritos individualmente em detalhes.</p>
    <p>
      <bold><italic>Plot</italic> de Poincar&#233;</bold>
    </p>
    <p>Um <italic>plot</italic> de Poincar&#233; &#233; um gr&#225;fico de dispers&#227;o de <italic>RR</italic><sub>n</sub> vs. <italic>RR</italic><sub>n+1</sub> onde <italic>RR</italic><sub>n</sub> &#233; o tempo entre dois picos R sucessivos e <italic>RR</italic><sub>n+1</sub> &#233; o tempo entre os dois picos R sucessivos seguintes. Quando o <italic>plot</italic> &#233; ajustado pelo m&#233;todo de ajuste por elipses, a an&#225;lise fornece tr&#234;s &#237;ndices: o desvio-padr&#227;o da variabilidade instant&#226;nea do intervalo batimento a batimento (SD1), a variabilidade de longo prazo dos intervalos R-R cont&#237;nuos (SD2), e a raz&#227;o SD1/SD2 (SD12)<sup>15</sup>. No <italic>plot</italic> de Poincar&#233;, SD1 &#233; a largura e SD2 o comprimento da elipse. Al&#233;m desse <italic>plot</italic> convencional (<italic>RR</italic><sub>n+1</sub> vs. <italic>RR</italic><sub>n</sub>), tamb&#233;m usamos o plot de Poincar&#233; generalizado com diferentes intervalos, incluindo o <italic>plot</italic> de Poincar&#233; m-lagged (o <italic>plot</italic> de <italic>RR</italic><sub>n+m</sub><italic>versus RR</italic><sub>n</sub>). Os valores de SD1 e SD2 foram calculados para lag = m a partir das rela&#231;&#245;es</p>
    <p>SD1 = (&#934;(m) - &#934;(0))<sup>1/2</sup> and SD2 = (&#934;(m) + &#934;(0))<sup>1/2</sup>, onde a fun&#231;&#227;o de autocovari&#226;ncia &#934; (m) &#233; dada por</p>
    <p>
      <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e01.jpg"/></p>
    <p>e <inline-formula id="car01"><inline-graphic xlink:href="https://minio.scielo.br/documentstore/1678-4170/zwb6ZdCrX9ZSDLBnb4fL58C/5dd9330853ebcfd958d3ea994acc6ca7e33bcaf7.jpg"/> &#233; a m&#233;dia de RR</inline-formula><sub>n</sub><sup>14</sup>. Para o prop&#243;sito do nosso estudo, definimos m em 1, 5 e 9. Em seguida, estendemos nossa an&#225;lise para revelar a associa&#231;&#227;o entre esses valores de desvio padr&#227;o (SD) e m usando a aproxima&#231;&#227;o de Pad&#233;<sup>19</sup>. Partimos do pressuposto de uma forma simples da aproxima&#231;&#227;o de Pad&#233; para valores de DP como a raz&#227;o polinomial em M de grau um.</p>
    <p>
      <disp-formula id="equ01">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e02.jpg"/></disp-formula>
    </p>
    <p>Aqui, Y = SD1, SD2, ou SD12 e &#967; = a / c. Os termos &#946; = b / a e &#947; = d / c s&#227;o os novos par&#226;metros desconhecidos. A fim de determinar se esses par&#226;metros t&#234;m valor na avalia&#231;&#227;o da sa&#250;de cardiovascular, consideramos a eq. (1) para o caso de "m". Nesse limite, a equa&#231;&#227;o (1) pode ser aproximada para Y = C + LM + QM<sup>2</sup>, onde a inclina&#231;&#227;o &#233; L = &#967; (&#946; - &#947;) e a curvatura &#233; Q = &#947;L . A inclina&#231;&#227;o e curvatura do <italic>plot</italic> de DP vs. m foram determinados pelos par&#226;metros ajustados &#967;, &#946; e &#947;.</p>
    <p>
      <bold>An&#225;lise de Flutua&#231;&#227;o Destendenciada</bold>
    </p>
    <p>Outro m&#233;todo anal&#237;tico para avaliar a correla&#231;&#227;o de longo prazo na sequ&#234;ncia de tempo R-R &#233; baseado na AFD<sup>20</sup>. A medida da correla&#231;&#227;o &#233; dada pelo expoente de escala (&#945;) da fun&#231;&#227;o de flutua&#231;&#227;o F(&#964;) &#8776; &#964;<sup>&#945;</sup>. A fun&#231;&#227;o de flutua&#231;&#227;o F (&#964;) foi calculada como se segue. Para uma dada sequ&#234;ncia de tempo R(t<sub>i</sub>), t<sub>i</sub> = i&#948;t, onde &#948;t &#233; o intervalo de tempo caracter&#237;stico para a sequ&#234;ncia e I = 1, N &#233; uma s&#233;rie de tempos integrados, r(t<sub>i</sub>) foi definido como r(t<sub>i</sub>) = &#931;<sup>i</sup><sub>j</sub> [R(t<sub>j</sub>) - &lt;R&gt;], I = 1,N, onde &lt;R&gt; &#233; a m&#233;dia de R(t<sub>i</sub>). A s&#233;rie integrada foi dividida em segmentos de igual dura&#231;&#227;o, &#964; = n &#948;t e uma fun&#231;&#227;o linear utilizada para ajustar os dados em cada segmento. A fun&#231;&#227;o de flutua&#231;&#227;o F (&#964;) foi calculada como valor quadr&#225;tico m&#233;dio da flutua&#231;&#227;o em rela&#231;&#227;o &#224; tend&#234;ncia linear, e o alfa foi obtido por ajuste dos dados a uma fun&#231;&#227;o de lei de pot&#234;ncia. Foi observado que uma estimativa aceit&#225;vel do dimensionamento do expoente alfa (a partir da AFD) pode ser obtida a partir da an&#225;lise dos conjuntos de dados, com 256 ou mais amostras (equivalente a cerca de 3,5 min de dados RR em uma frequ&#234;ncia card&#237;aca de 70 batimentos/min). Portanto, era esperado que a an&#225;lise dos dados RR de um per&#237;odo de grava&#231;&#227;o de ECG de 10 min, fornecesse uma medida adequada do expoente de escala<sup>21</sup>. No entanto, o valor de alfa obtido a partir desse c&#225;lculo pode estar sob influ&#234;ncia tanto da escala de curto prazo, refletindo o controle parassimp&#225;tico, quanto da escala de longo prazo, refletindo o controle simp&#225;tico e, portanto, pode falhar em distinguir por completo as influ&#234;ncias parassimp&#225;tica e simp&#225;tica. Uma an&#225;lise separada tanto da escala de curto prazo quanto da de longo prazo deve anular o efeito m&#250;tuo e revelar a varia&#231;&#227;o exata da escala<sup>22</sup>. Assim, foram analisados os valores separados de alfa, &#945;<sub>s</sub> de curto prazo e &#945;<sub>l</sub> de longo prazo. Para o &#945;<sub>s </sub>foram inclu&#237;dos 25 batimentos, enquanto para o &#945;<sub>l </sub>foram inclu&#237;dos dados de 30 a N/4 batimentos. </p>
    <p>
      <bold>Correla&#231;&#227;o entre as diferen&#231;as sucessivas do intervalo RR<sub>n</sub></bold>
    </p>
    <p>A coer&#234;ncia do intervalo RR<sub>n</sub> pode ser avaliada a partir do mapa de varia&#231;&#227;o do intervalo:</p>
    <p>
      <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e03.jpg"/></p>
    <p>onde &lt;RR<sub>n</sub>&gt; &#233; o intervalo m&#233;dio. Espera-se que esse <italic>plot </italic>mostre a correla&#231;&#227;o entre a variabilidade dos tr&#234;s intervalos R-R consecutivos. </p>
    <p>
      <bold>Autocorrela&#231;&#227;o da flutua&#231;&#227;o de RR<sub>n</sub></bold>
    </p>
    <p>Exploramos a autocorrela&#231;&#227;o do desvio de RR<sub>n</sub> da m&#233;dia &lt;RR<sub>n</sub>&gt;. A fun&#231;&#227;o de autocorrela&#231;&#227;o C(m) de um indiv&#237;duo em particular foi calculada como</p>
    <p>
      <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e04.jpg"/></p>
    <p>onde o desvio &#233; &#8710;RR<sub>n </sub>= RR<sub>n </sub>- (RR<sub>n</sub>) e N &#233; o n&#250;mero total de intervalos RR<sub>n</sub>.</p>
    <p>
      <bold>Resultados</bold>
    </p>
    <p>A frequ&#234;ncia card&#237;aca m&#233;dia foi de 74,7 &#144;&#177; 6,1 batimentos/min para o grupo de diab&#233;ticos e 72,4 &#144;&#177; 6,7 batimentos/min para o grupo controle saud&#225;vel. A m&#233;dia de idade do grupo de diab&#233;ticos foi de 46,3 anos (varia&#231;&#227;o, 36-56 anos), e no grupo de controle foi de 47,4 anos (varia&#231;&#227;o, 39-57 anos). Todos os indiv&#237;duos do estudo eram normotensos.</p>
    <p>Na an&#225;lise do <italic>plot</italic> de Poincar&#233;, o gr&#225;fico de dispers&#227;o aumentou com o n&#250;mero de lag, gerando maior largura (SD1) e comprimento (SD2). O aumento incremental na largura do <italic>plot RR</italic><sub>n+m</sub> vs. <italic>RR</italic><sub>n</sub> &#224; medida que m aumentava foi menor no grupo de diabetes (<xref ref-type="fig" rid="aop_5014fig01">Figura 1</xref>, D) do que no grupo controle (<xref ref-type="fig" rid="aop_5014fig01">Figura 1</xref>, D). As diferen&#231;as nos valores de SD1, SD2, e SD12 entre o grupo de diabetes e o grupo controle foram estatisticamente significativas (p &lt; 0,001 para todos). Os valores de SD1 e SD12 foram mais elevados no grupo controle, enquanto o SD2 foi maior no grupo diab&#233;tico. A diferen&#231;a de SD12 aumentou com o n&#250;mero de lag (<xref ref-type="fig" rid="aop_5014fig02">Figura 2</xref>).</p>
    <p content-type="asset">
      <fig id="aop_5014fig01">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf01.jpg"/></fig>
    </p>
    <p content-type="asset">
      <fig id="aop_5014fig02">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf02.jpg"/></fig>
    </p>
    <p>Um excelente ajuste dos dados com a equa&#231;&#227;o (1) (linha s&#243;lida na curva, R<sup>2</sup> = 0,999), foi encontrado com o conjunto de dados &#967;, &#946;, &#947; listados na <xref ref-type="table" rid="aop_5014tab01">Tabela 1</xref>. Os valores de L e Q tal como obtidos pelo ajuste dos dados com a eq. (1) tamb&#233;m s&#227;o apresentados na <xref ref-type="table" rid="aop_5014tab01">Tabela 1</xref>. As caracter&#237;sticas gerais mostraram que a inclina&#231;&#227;o (L) foi positiva, mas a curvatura (Q) negativa para todos os par&#226;metros e a curvatura foi quase uma ordem de grandeza menor do que a inclina&#231;&#227;o.</p>
    <p content-type="asset">
      <table-wrap id="aop_5014tab01">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gt01.jpg"/></table-wrap>
    </p>
    <p>A partir da AFD, o valor m&#233;dio de alfa para o grupo controle foi menor do que para o grupo de diab&#233;ticos (0,88 &#144;&#177; 0,17 vs. 1,02 &#144;&#177; 0,13, p &lt;0,001) (<xref ref-type="fig" rid="aop_5014fig03">Figura 3</xref>). Em indiv&#237;duos controle, &#945;<sub>s</sub> foi ligeiramente maior do que &#945;<sub>l</sub> (1,01 &#144;&#177; 0,14 vs 0,80 &#144;&#177; 0,19), enquanto &#945;<sub>l</sub> foi maior do que &#945;<sub>s</sub> para o grupo diab&#233;tico (&#945;<sub>s</sub> = 1,09 &#144;&#177; 0,17; &#945;<sub>l</sub> = 1,18 &#144;&#177; 0,19). Quando &#945;<sub>s</sub> foi plotado contra &#945;<sub>l</sub> (<xref ref-type="fig" rid="aop_5014fig04">Figura 4</xref>), a popula&#231;&#227;o diab&#233;tica e n&#227;o-diab&#233;tica tenderam a formar dois grupos separados.</p>
    <p content-type="asset">
      <fig id="aop_5014fig03">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf03.jpg"/></fig>
    </p>
    <p content-type="asset">
      <fig id="aop_5014fig04">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf04.jpg"/></fig>
    </p>
    <p>No <italic>plot</italic> de correla&#231;&#227;o, os pontos estavam agrupados em torno da origem para pacientes diab&#233;ticos. Em contrapartida, houve uma maior dispers&#227;o sobre a origem e mais assimetria no <italic>plot</italic> de indiv&#237;duos controle (<xref ref-type="fig" rid="aop_5014fig05">Figura 5</xref>, ND1, ND2). A for&#231;a da correla&#231;&#227;o do ritmo card&#237;aco foi estimada considerando a autocorrela&#231;&#227;o de flutua&#231;&#227;o no RR<sub>n</sub>. Na <xref ref-type="fig" rid="aop_5014fig06">Figura 6</xref>, s&#227;o plotados resultados representativos de um paciente controle e de um paciente diab&#233;tico. As fun&#231;&#245;es de autocorrela&#231;&#227;o para pacientes diab&#233;ticos e controle foram distintas. Para diab&#233;ticos, a fun&#231;&#227;o de correla&#231;&#227;o C(m) diminuiu lentamente (curva preta e verde na figura superior), com tempo de lag. A depend&#234;ncia do tempo ficou pr&#243;xima da soma dos dois exponenciais com pequena amplitude sobreposta de oscila&#231;&#227;o de baixa frequ&#234;ncia. Por outro lado, a C(m) dos indiv&#237;duos saud&#225;veis demonstrou uma queda mais r&#225;pida (exponencial) com a diminui&#231;&#227;o do tempo de correla&#231;&#227;o em compara&#231;&#227;o com os casos de diabetes. Para confirmar essa diferen&#231;a no padr&#227;o de correla&#231;&#227;o entre os indiv&#237;duos controle e diab&#233;ticos, rearranjamos a s&#233;rie de tempo real do intervalo R-R usando o software Matlab e as fun&#231;&#245;es de autocorrela&#231;&#227;o dos dados rearranjados (vermelho e azul para indiv&#237;duos 1 e 2, respectivamente) foram representados na <xref ref-type="fig" rid="aop_5014fig06">Figura 6</xref>. As fun&#231;&#245;es de autocorrela&#231;&#227;o dos dados rearranjados de todos os indiv&#237;duos (dois diab&#233;ticos e dois controles saud&#225;veis) foram quase id&#234;nticas.</p>
    <p content-type="asset">
      <fig id="aop_5014fig05">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf05.jpg"/></fig>
    </p>
    <p content-type="asset">
      <fig id="aop_5014fig06">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf06.jpg"/></fig>
    </p>
    <p>Tamb&#233;m caracterizamos as propriedades da &#8710;RRn pela fun&#231;&#227;o de distribui&#231;&#227;o de probabilidade P (&#8710;RRn) (<xref ref-type="fig" rid="aop_5014fig07">Figura 7</xref>). Para os pacientes diab&#233;ticos, a distribui&#231;&#227;o de probabilidade foi quase sim&#233;trica e poderia ser ajustada por uma fun&#231;&#227;o de Gauss (R2 = 0,93) com largura = 0,023. Para indiv&#237;duos saud&#225;veis, a distribui&#231;&#227;o de probabilidade P foi assim&#233;trica com m&#233;dia positiva e largura maior = 0,036, como obtida atrav&#233;s do ajuste de Gauss (R<sup>2</sup> = 0,93).</p>
    <p content-type="asset">
      <fig id="aop_5014fig07">
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf07.jpg"/></fig>
    </p>
    <p>
      <bold>Discuss&#227;o</bold>
    </p>
    <p>Encontramos diferen&#231;as marcantes no padr&#227;o de VFC entre indiv&#237;duos saud&#225;veis de controle e indiv&#237;duos diab&#233;ticos por meio de an&#225;lises n&#227;o-lineares. Os pacientes foram pareados por idade m&#233;dia e frequ&#234;ncia card&#237;aca de repouso, os dois principais determinantes da HRV<sup>23</sup>, de modo que a diferen&#231;a na distribui&#231;&#227;o iria refletir as mudan&#231;as na regula&#231;&#227;o cardiovascular resultantes apenas da condi&#231;&#227;o diab&#233;tica. </p>
    <p>V&#225;rias modifica&#231;&#245;es do <italic>plot</italic> simples de Poincar&#233; t&#234;m sido propostas para revelar mais eficazmente as mudan&#231;as nos padr&#245;es de HRV, incluindo o <italic>plot</italic> lagged. O conceito desse <italic>plot</italic> lagged m surgiu do reconhecimento de que qualquer intervalo R-R pode influenciar at&#233; oito intervalos R-R subsequentes<sup>24,25</sup>. Tem sido demonstrado que SD1 est&#225; correlacionado com a variabilidade de curto prazo da frequ&#234;ncia card&#237;aca e &#233; influenciado principalmente pela modula&#231;&#227;o parassimp&#225;tica, enquanto SD2 &#233; uma medida da variabilidade de longo prazo<sup>14,26</sup> e reflete a ativa&#231;&#227;o simp&#225;tica. O SD1 menor em indiv&#237;duos diab&#233;ticos indica que a regula&#231;&#227;o parassimp&#225;tica est&#225; enfraquecida pela doen&#231;a, presumivelmente por neuropatia perif&#233;rica, enquanto que um SD2 maior em pacientes diab&#233;ticos indica aumento da variabilidade de longo prazo devido ao <italic>input</italic> compensat&#243;rio simp&#225;tico. </p>
    <p>Os resultados da an&#225;lise do <italic>plot</italic> de Poincar&#233; s&#227;o demonstrados pela inclina&#231;&#227;o (L) e curvatura (-Q) do <italic>plot</italic>. No grupo diab&#233;tico, L e -Q para SD1 e SD12 foram menores, enquanto que os valores de L e -Q para SD2 foram maiores do que no grupo controle. A diferen&#231;a em Q foi maior do que a diferen&#231;a em L. Em particular, o valor de Q para SD12 no grupo controle foi &gt; 3 vezes maior do que no grupo diab&#233;tico. Baixos valores de curvatura s&#227;o encontrados em pacientes com doen&#231;as cardiovasculares<sup>24</sup>. Esses dados sugerem fortemente a diminui&#231;&#227;o da atividade parassimp&#225;tica e influ&#234;ncia excessiva da atividade simp&#225;tica no cora&#231;&#227;o diab&#233;tico. Al&#233;m disso, esse resultado fornece apoio indireto para a no&#231;&#227;o de que a maior influ&#234;ncia simp&#225;tica sobre a fun&#231;&#227;o cardiovascular est&#225; correlacionada com a morbidade card&#237;aca<sup>27,28</sup>. Um SD12 aumentado &#233; considerado um bom indicador da din&#226;mica de um cora&#231;&#227;o saud&#225;vel, e o valor mais baixo em pacientes diab&#233;ticos novamente sustenta um balan&#231;o simp&#225;tico-vagal alterado em diab&#233;ticos.</p>
    <p>Estudos anteriores usando AFD mostraram que &#945;<sub>s</sub> &gt; &#945;<sub>1</sub> em indiv&#237;duos saud&#225;veis, enquanto que o inverso ocorreu em indiv&#237;duos com doen&#231;a cardiovascular<sup>20</sup>. Encontramos uma tend&#234;ncia semelhante no presente estudo, novamente confirmando o efeito adverso da diabetes no cora&#231;&#227;o.</p>
    <p>Na aus&#234;ncia de modula&#231;&#227;o externa, espera-se que o plot de correla&#231;&#227;o disperse-se pr&#243;ximo do ponto de origem, enquanto o <italic>input</italic> aleat&#243;rio produz uma distribui&#231;&#227;o uniforme. Observamos uma alta densidade de pontos em torno da origem com maior simetria em pacientes diab&#233;ticos em compara&#231;&#227;o com os controles. <italic>Plots</italic> de controles saud&#225;veis eram de forma geral dispersos assimetricamente com grandes valores de RR<sub>n</sub>. Esses resultados sugerem que os mecanismos de desacelera&#231;&#227;o e acelera&#231;&#227;o de FC em diferentes per&#237;odos de tempo est&#227;o substancialmente prejudicados em pacientes diab&#233;ticos.</p>
    <p>A aplica&#231;&#227;o da autocorrela&#231;&#227;o &#224; an&#225;lise da VFC &#233; uma id&#233;ia recente que considera a VFC como o resultado da intera&#231;&#227;o entre os osciladores acoplados de v&#225;rias frequ&#234;ncias<sup>29</sup>. O grau de autocorrela&#231;&#227;o pode tamb&#233;m refletir sobre as escalas de tempo incorporadas dentro do padr&#227;o da VFC. Considera-se que cada uma dessas escalas de tempo do oscilador acoplado seja representada por uma intera&#231;&#227;o do auto-oscilador isolado, interagindo com outros osciladores com diferentes fun&#231;&#245;es fisiol&#243;gicas<sup>18</sup>. A falta de queda exponencial em C(m) indica a presen&#231;a de um efeito de mem&#243;ria de longo prazo do estado diab&#233;tico e sugere fortemente que os mecanismos para a varia&#231;&#227;o de curto prazo na frequ&#234;ncia card&#237;aca est&#227;o enfraquecidos ou ausentes nos pacientes diab&#233;ticos.</p>
    <p>A an&#225;lise da variabilidade da frequ&#234;ncia card&#237;aca baseada na din&#226;mica n&#227;o-linear tem se mostrado superior aos m&#233;todos convencionais para a identifica&#231;&#227;o de altera&#231;&#245;es ocultas na modula&#231;&#227;o auton&#244;mica card&#237;aca em v&#225;rias condi&#231;&#245;es de doen&#231;a. Estudos anteriores demonstraram diferen&#231;as nos <italic>plots</italic> de Poincar&#233;, AFD e an&#225;lises de amostra de entropia entre os cora&#231;&#245;es de pacientes diab&#233;ticos e n&#227;o diab&#233;ticos<sup>17,30</sup>, embora muitas vezes essas diferen&#231;as n&#227;o alcan&#231;assem signific&#226;ncia estat&#237;stica devido aos pequenos tamanhos das amostras<sup>17</sup>. Nosso estudo n&#227;o s&#243; analisou um n&#250;mero maior de pacientes e controles, mas tamb&#233;m usou v&#225;rias ferramentas anal&#237;ticas n&#227;o-lineares, incluindo an&#225;lise de <italic>plot </italic>de Poincar&#233;, AFD e an&#225;lise de autocorrela&#231;&#227;o para demonstrar as altera&#231;&#245;es na VFC causadas pela neuropatia diab&#233;tica.</p>
    <p>A principal limita&#231;&#227;o deste estudo &#233; a popula&#231;&#227;o heterog&#234;nea de pacientes. A dura&#231;&#227;o da doen&#231;a no grupo de pacientes foi vari&#225;vel e muitos estavam recebendo diferentes medicamentos antidiab&#233;ticos. Al&#233;m disso, uma popula&#231;&#227;o de 23 pacientes pode ser suficiente para identificar as diferen&#231;as no padr&#227;o de VFC entre indiv&#237;duos diab&#233;ticos e n&#227;o diab&#233;ticos saud&#225;veis, mas um grupo muito maior de pacientes &#233; necess&#225;rio para confirmar os verdadeiros valores de diagn&#243;stico e progn&#243;stico dos par&#226;metros derivados a partir dos m&#233;todos anal&#237;ticos. O pr&#243;ximo passo l&#243;gico &#233; a an&#225;lise intra-grupo em um grupo maior de pacientes diab&#233;ticos com dura&#231;&#227;o da doen&#231;a vari&#225;vel para avaliar as mudan&#231;as progressivas no padr&#227;o de VFC. Nosso estudo estabelece o potencial de m&#233;todos n&#227;o-lineares de an&#225;lise da variabilidade da frequ&#234;ncia card&#237;aca para avaliar mudan&#231;as no padr&#227;o de VFC indicativo de doen&#231;a cardiovascular, incluindo os efeitos associados com o diabetes mellitus.</p>
    <p>
      <bold>Conclus&#245;es</bold>
    </p>
    <p>Em resumo, n&#243;s mostramos a efic&#225;cia dos m&#233;todos de an&#225;lise n&#227;o-linear para estudar as diferen&#231;as no padr&#227;o de VFC entre pacientes diab&#233;ticos e controles pareados saud&#225;veis. Tamb&#233;m enfatizamos a novidade da an&#225;lise de autocorrela&#231;&#227;o para avaliar as mudan&#231;as na regula&#231;&#227;o auton&#244;mica do cora&#231;&#227;o diab&#233;tico. Que seja de nosso conhecimento, essa &#233; a primeira tentativa de distinguir a fun&#231;&#227;o card&#237;aca normal da fun&#231;&#227;o card&#237;aca diab&#233;tica atrav&#233;s da an&#225;lise de autocorrela&#231;&#227;o. Acreditamos que esses m&#233;todos t&#234;m o potencial para identificar marcadores de diagn&#243;stico e progn&#243;stico para a neuropatia auton&#244;mica card&#237;aca em pacientes com diabetes.</p>
    <p>
      <bold>Contribui&#231;&#227;o dos autores</bold>
    </p>
    <p>Concep&#231;&#227;o e desenho da pesquisa, Obten&#231;&#227;o de dados, An&#225;lise e interpreta&#231;&#227;o dos dados, An&#225;lise estat&#237;stica, Reda&#231;&#227;o do manuscrito e Revis&#227;o cr&#237;tica do manuscrito quanto ao conte&#250;do intelectual: Roy, B, Ghatak S.</p>
    <p>
      <bold>Potencial Conflito de Interesses</bold>
    </p>
    <p>Declaro n&#227;o haver conflito de interesses pertinentes.</p>
    <p>
      <bold>Fontes de Financiamento</bold>
    </p>
    <p>O presente estudo n&#227;o teve fontes de financiamento externas.</p>
    <p>
      <bold>Vincula&#231;&#227;o Acad&#234;mica</bold>
    </p>
    <p>N&#227;o h&#225; vincula&#231;&#227;o deste estudo a programas de p&#243;s-gradua&#231;&#227;o.</p>
    <p>
      <fn id="end" fn-type="corresp">
        <label>
          <bold>Correspond&#234;ncia:</bold>
        </label>
        <p> Bhaskar Roy </p>
        <p> 263 Farmington Avenue, Farmington, CT, 06030 </p>
        <p> E-mail: </p>
        <p><email xlink:href="imbhaskarall@gmail.com">imbhaskarall@gmail.com</email>, </p>
        <p>
          <email xlink:href="broy2@rics.bwh.harvard.edu">broy2@rics.bwh.harvard.edu</email>
        </p>
      </fn>
    </p>
    <p>Artigo Recebido em 17/07/12, Revisado em 09/09/12, aceito em 26/04/13.</p>
  </body>
  <back>
    <ref-list>
      <ref id="B1">
        <mixed-citation>1. Contreras P, Canetti R, Migliaro RE. Correlations between frequency-domain HRV indices and lagged Poincar&#233; plot width in healthy and diabetic subjects. Physiol Meas. 2007;28(1):85-94.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Correlations between frequency-domain HRV indices and lagged Poincar&#233; plot width in healthy and diabetic subjects</article-title>
          <source>Physiol Meas</source>
          <date>
            <year>2007</year>
          </date>
          <fpage>85</fpage>
          <lpage>94</lpage>
          <issue>1</issue>
          <volume>28</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Contreras</surname>
              <given-names>P</given-names>
            </name>
            <name>
              <surname>Canetti</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Migliaro</surname>
              <given-names>RE</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B2">
        <mixed-citation>2. Rajendra Acharya U, Paul Joseph K, Kannathal N, Lim CM, Suri JS. Heart rate variability: a review. Med Biol Eng Comput. 2006;44(12):1031-51.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Heart rate variability: a review</article-title>
          <source>Med Biol Eng Comput</source>
          <date>
            <year>2006</year>
          </date>
          <fpage>1031</fpage>
          <lpage>51</lpage>
          <issue>12</issue>
          <volume>44</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Rajendra Acharya</surname>
              <given-names>U</given-names>
            </name>
            <name>
              <surname>Paul Joseph</surname>
              <given-names>K</given-names>
            </name>
            <name>
              <surname>Kannathal</surname>
              <given-names>N</given-names>
            </name>
            <name>
              <surname>Lim</surname>
              <given-names>CM</given-names>
            </name>
            <name>
              <surname>Suri</surname>
              <given-names>JS</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B3">
        <mixed-citation>3. Tsuji H, Larson MG, Venditti FJ Jr, Manders ES, Evans JC, Feldman CL, et al. Impact of reduced heart rate variability on risk for cardiac events. The Framingham Heart Study. Circulation. 1996;94(11):2850-5.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Impact of reduced heart rate variability on risk for cardiac events: The Framingham Heart Study</article-title>
          <source>Circulation</source>
          <date>
            <year>1996</year>
          </date>
          <fpage>2850</fpage>
          <lpage>5</lpage>
          <issue>11</issue>
          <volume>94</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Tsuji</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Larson</surname>
              <given-names>MG</given-names>
            </name>
            <name>
              <surname>Venditti Jr</surname>
              <given-names>FJ</given-names>
            </name>
            <name>
              <surname>Manders</surname>
              <given-names>ES</given-names>
            </name>
            <name>
              <surname>Evans</surname>
              <given-names>JC</given-names>
            </name>
            <name>
              <surname>Feldman</surname>
              <given-names>CL</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B4">
        <mixed-citation>4. Vinik IA, Maser RE, Mitchell BD, Freeman R. Diabetic autonomic neuropathy. Diabetes Care. 2003;26(5):1553-79.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Diabetic autonomic neuropathy</article-title>
          <source>Diabetes Care</source>
          <date>
            <year>2003</year>
          </date>
          <fpage>1553</fpage>
          <lpage>79</lpage>
          <issue>5</issue>
          <volume>26</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Vinik</surname>
              <given-names>IA</given-names>
            </name>
            <name>
              <surname>Maser</surname>
              <given-names>RE</given-names>
            </name>
            <name>
              <surname>Mitchell</surname>
              <given-names>BD</given-names>
            </name>
            <name>
              <surname>Freeman</surname>
              <given-names>R</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B5">
        <mixed-citation>5. Gerritsen J, Dekker JM, TenVoorde BJ, Kostense PJ, Heine RJ, Bouter LM, et al. Impaired autonomic function is associated with increased mortality especially in subjects with diabetes, hypertension, or a history of cardiovascular disease: the Hoorn Study. Diabetes Care. 2001;24(10):1793-8.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Impaired autonomic function is associated with increased mortality especially in subjects with diabetes, hypertension, or a history of cardiovascular disease: the Hoorn Study</article-title>
          <source>Diabetes Care</source>
          <date>
            <year>2001</year>
          </date>
          <fpage>1793</fpage>
          <lpage>8</lpage>
          <issue>10</issue>
          <volume>24</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Gerritsen</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Dekker</surname>
              <given-names>JM</given-names>
            </name>
            <name>
              <surname>TenVoorde</surname>
              <given-names>BJ</given-names>
            </name>
            <name>
              <surname>Kostense</surname>
              <given-names>PJ</given-names>
            </name>
            <name>
              <surname>Heine</surname>
              <given-names>RJ</given-names>
            </name>
            <name>
              <surname>Bouter</surname>
              <given-names>LM</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B6">
        <mixed-citation>6. Liao D, Carnethon M, Evans GW, Cascio WE, Heiss G. Lower heart rate variability is associated with the development of coronary heart disease in individuals with diabetes. The Atherosclerosis Risk in Communities (ARIC) Study. Diabetes. 2002;51(12):3524-31.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Lower heart rate variability is associated with the development of coronary heart disease in individuals with diabetes: The Atherosclerosis Risk in Communities (ARIC) Study</article-title>
          <source>Diabetes</source>
          <date>
            <year>2002</year>
          </date>
          <fpage>3524</fpage>
          <lpage>31</lpage>
          <issue>12</issue>
          <volume>51</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Liao</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Carnethon</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Evans</surname>
              <given-names>GW</given-names>
            </name>
            <name>
              <surname>Cascio</surname>
              <given-names>WE</given-names>
            </name>
            <name>
              <surname>Heiss</surname>
              <given-names>G</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B7">
        <mixed-citation>7. Kataoka M, Ito C, Sasaki H, Yamaneb K, Kohno N. Low heart rate variability is a risk factor for sudden cardiac death in type 2 diabetes. Diabetes Res Clin Pract. 2004;64(1):51-8.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Low heart rate variability is a risk factor for sudden cardiac death in type 2 diabetes</article-title>
          <source>Diabetes Res Clin Pract</source>
          <date>
            <year>2004</year>
          </date>
          <fpage>51</fpage>
          <lpage>8</lpage>
          <issue>1</issue>
          <volume>64</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Kataoka</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Ito</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Sasaki</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Yamaneb</surname>
              <given-names>K</given-names>
            </name>
            <name>
              <surname>Kohno</surname>
              <given-names>N</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B8">
        <mixed-citation>8. Alina JK, Agata MG, Torzynska K, Kramer L, Sowinska A, Moczko J, et al. Diabetes abolishes the influence of revascularization on heart rate variability in patients with stable angina, Assessment by novel mathematical models. [abstract]. J Electrocardiol. 2007;40:S32-S33.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Diabetes abolishes the influence of revascularization on heart rate variability in patients with stable angina, Assessment by novel mathematical models</article-title>
          <source>J Electrocardiol</source>
          <date>
            <year>2007</year>
          </date>
          <fpage>S32</fpage>
          <lpage>S33</lpage>
          <volume>40</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Alina</surname>
              <given-names>JK</given-names>
            </name>
            <name>
              <surname>Agata</surname>
              <given-names>MG</given-names>
            </name>
            <name>
              <surname>Torzynska</surname>
              <given-names>K</given-names>
            </name>
            <name>
              <surname>Kramer</surname>
              <given-names>L</given-names>
            </name>
            <name>
              <surname>Sowinska</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Moczko</surname>
              <given-names>J</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B9">
        <mixed-citation>9. Ewing DJ, Martyn CM, Young RJ, Clarke BF. The value of cardiovascular autonomic function tests: 10 years experience in diabetes. Diabetes Care. 1985;8(5):491-8.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>The value of cardiovascular autonomic function tests: 10 years experience in diabetes</article-title>
          <source>Diabetes Care</source>
          <date>
            <year>1985</year>
          </date>
          <fpage>491</fpage>
          <lpage>8</lpage>
          <issue>5</issue>
          <volume>8</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Ewing</surname>
              <given-names>DJ</given-names>
            </name>
            <name>
              <surname>Martyn</surname>
              <given-names>CM</given-names>
            </name>
            <name>
              <surname>Young</surname>
              <given-names>RJ</given-names>
            </name>
            <name>
              <surname>Clarke</surname>
              <given-names>BF</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B10">
        <mixed-citation>10. Pagani M. Heart rate variability and autonomic diabetic neuropathy. Diabetes Nutr Metab. 2000;13(6):341-6.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Heart rate variability and autonomic diabetic neuropathy</article-title>
          <source>Diabetes Nutr Metab</source>
          <date>
            <year>2000</year>
          </date>
          <fpage>341</fpage>
          <lpage>6</lpage>
          <issue>6</issue>
          <volume>13</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Pagani</surname>
              <given-names>M</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B11">
        <mixed-citation>11. Rolim LC, Sa JR, Chacra AR, Dib SA. Diabetic cardiovascular autonomic neuropathy: risk factors, clinical impact and early diagnosis. Arq Bras Cardiol. 2008;90(4):e24-31.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Diabetic cardiovascular autonomic neuropathy: risk factors, clinical impact and early diagnosis</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>2008</year>
          </date>
          <fpage>e24</fpage>
          <lpage>31</lpage>
          <issue>4</issue>
          <volume>90</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Rolim</surname>
              <given-names>LC</given-names>
            </name>
            <name>
              <surname>Sa</surname>
              <given-names>JR</given-names>
            </name>
            <name>
              <surname>Chacra</surname>
              <given-names>AR</given-names>
            </name>
            <name>
              <surname>Dib</surname>
              <given-names>SA</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B12">
        <mixed-citation>12. Spallone V, Menzinger G. Diagnosis of cardiovascular autonomic neuropathy in diabetes. Diabetes. 1997;46 Suppl 2:S67-76.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Diagnosis of cardiovascular autonomic neuropathy in diabetes</article-title>
          <source>Diabetes</source>
          <date>
            <year>1997</year>
          </date>
          <fpage>S67</fpage>
          <lpage>76</lpage>
          <volume>46</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Spallone</surname>
              <given-names>V</given-names>
            </name>
            <name>
              <surname>Menzinger</surname>
              <given-names>G</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B13">
        <mixed-citation>13. Tulppo MP, Makikallio TH, Takala TE, Seppanen T, Huikuri HV. Quantitative beat-to-beat analysis of heart rate dynamics during exercise. Am J Physiol. 1996;271(1 Pt 2):H244-52.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Quantitative beat-to-beat analysis of heart rate dynamics during exercise</article-title>
          <source>Am J Physiol</source>
          <date>
            <year>1996</year>
          </date>
          <fpage>H244</fpage>
          <lpage>52</lpage>
          <issue>1</issue>
          <volume>271</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Tulppo</surname>
              <given-names>MP</given-names>
            </name>
            <name>
              <surname>Makikallio</surname>
              <given-names>TH</given-names>
            </name>
            <name>
              <surname>Takala</surname>
              <given-names>TE</given-names>
            </name>
            <name>
              <surname>Seppanen</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Huikuri</surname>
              <given-names>HV</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B14">
        <mixed-citation>14. Brennan M, Palaniswami M, Kamen P. Do existing measures of Poincar&#233; plot geometry reflect nonlinear features of heart rate variability? IEEE Trans Biomed Eng. 2001;48(11):1342-7.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Do existing measures of Poincar&#233; plot geometry reflect nonlinear features of heart rate variability?</article-title>
          <source>IEEE Trans Biomed Eng</source>
          <date>
            <year>2001</year>
          </date>
          <fpage>1342</fpage>
          <lpage>7</lpage>
          <issue>11</issue>
          <volume>48</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Brennan</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Palaniswami</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Kamen</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B15">
        <mixed-citation>15. Woo MA, Stevenson WG, Moser DK, Trelease RB, Harper RM. Patterns of beat-to-beat heart rate variability in advanced heart failure. Am Heart J. 1992;123(3):704-10.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Patterns of beat-to-beat heart rate variability in advanced heart failure</article-title>
          <source>Am Heart J</source>
          <date>
            <year>1992</year>
          </date>
          <fpage>704</fpage>
          <lpage>10</lpage>
          <issue>3</issue>
          <volume>123</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Woo</surname>
              <given-names>MA</given-names>
            </name>
            <name>
              <surname>Stevenson</surname>
              <given-names>WG</given-names>
            </name>
            <name>
              <surname>Moser</surname>
              <given-names>DK</given-names>
            </name>
            <name>
              <surname>Trelease</surname>
              <given-names>RB</given-names>
            </name>
            <name>
              <surname>Harper</surname>
              <given-names>RM</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B16">
        <mixed-citation>16. Khandokera AH, Jelinekb HF, Moritani T, Palaniswami M. Association of cardiac autonomic neuropathy with alteration of sympatho-vagal balance through heart rate variability analysis. Med Eng Phys. 2010;32(2):161-7.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Association of cardiac autonomic neuropathy with alteration of sympatho-vagal balance through heart rate variability analysis</article-title>
          <source>Med Eng Phys</source>
          <date>
            <year>2010</year>
          </date>
          <fpage>161</fpage>
          <lpage>7</lpage>
          <issue>2</issue>
          <volume>32</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Khandokera</surname>
              <given-names>AH</given-names>
            </name>
            <name>
              <surname>Jelinekb</surname>
              <given-names>HF</given-names>
            </name>
            <name>
              <surname>Moritani</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Palaniswami</surname>
              <given-names>M</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B17">
        <mixed-citation>17. Khandoker AH, Jelinek HF, Palaniswami M. Identifying diabetic patients with cardiac autonomic neuropathy by heart rate complexity analysis. Biomed Eng Online. 2009;8:3.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Identifying diabetic patients with cardiac autonomic neuropathy by heart rate complexity analysis</article-title>
          <source>Biomed Eng Online</source>
          <date>
            <year>2009</year>
          </date>
          <fpage>3</fpage>
          <volume>8</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Khandoker</surname>
              <given-names>AH</given-names>
            </name>
            <name>
              <surname>Jelinek</surname>
              <given-names>HF</given-names>
            </name>
            <name>
              <surname>Palaniswami</surname>
              <given-names>M</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B18">
        <mixed-citation>18. Khovanov IA, Khovanova NA, McClintock PV, Stefanovska A. Intrinsic dynamics of heart regulatory systems on short time-scales: from experiment to modeling. [Cited on 2012 Jan 10]. Available from: <ext-link ext-link-type="uri" xlink:href="http://arxiv:org/PScache/arxov/pdf/0912/0912:2237v1.pdf">http://arxiv:org/PScache/arxov/pdf/0912/0912:2237v1.pdf</ext-link>.</mixed-citation>
        <element-citation publication-type="book">
          <source>Intrinsic dynamics of heart regulatory systems on short time-scales: from experiment to modeling</source>
          <person-group person-group-type="author">
            <name>
              <surname>Khovanov</surname>
              <given-names>IA</given-names>
            </name>
            <name>
              <surname>Khovanova</surname>
              <given-names>NA</given-names>
            </name>
            <name>
              <surname>McClintock</surname>
              <given-names>PV</given-names>
            </name>
            <name>
              <surname>Stefanovska</surname>
              <given-names>A</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B19">
        <mixed-citation>19. Ghatak SK, Roy B, Choudhuri R, Bandopadhaya R. Modulation of autonomous nervous system activity by gyrosound stimulation; 2010. [Cited on 2012 Feb 20]. Available from: <ext-link ext-link-type="uri" xlink:href="http://arxiv:org/abs/1003.2075">http://arxiv:org/abs/1003.2075</ext-link></mixed-citation>
        <element-citation publication-type="book">
          <source>Modulation of autonomous nervous system activity by gyrosound stimulation</source>
          <date>
            <year>2010</year>
          </date>
          <person-group person-group-type="author">
            <name>
              <surname>Ghatak</surname>
              <given-names>SK</given-names>
            </name>
            <name>
              <surname>Roy</surname>
              <given-names>B</given-names>
            </name>
            <name>
              <surname>Choudhuri</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Bandopadhaya</surname>
              <given-names>R</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B20">
        <mixed-citation>20. Peng CK, Havlin S, Stanley HE, Goldberger AL. Quantication of scaling exponents and crossover phenomena in nonstationary heartbeat time series. Chaos. 1995;5(1):82-7.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Quantication of scaling exponents and crossover phenomena in nonstationary heartbeat time series</article-title>
          <source>Chaos</source>
          <date>
            <year>1995</year>
          </date>
          <fpage>82</fpage>
          <lpage>7</lpage>
          <issue>1</issue>
          <volume>5</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Peng</surname>
              <given-names>CK</given-names>
            </name>
            <name>
              <surname>Havlin</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Stanley</surname>
              <given-names>HE</given-names>
            </name>
            <name>
              <surname>Goldberger</surname>
              <given-names>AL</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B21">
        <mixed-citation>21. Coronado AV, Carpena P. Size effects on correlation measures. J Biol Phys. 2005;31(1):121-33.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Size effects on correlation measures</article-title>
          <source>J Biol Phys</source>
          <date>
            <year>2005</year>
          </date>
          <fpage>121</fpage>
          <lpage>33</lpage>
          <issue>1</issue>
          <volume>31</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Coronado</surname>
              <given-names>AV</given-names>
            </name>
            <name>
              <surname>Carpena</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B22">
        <mixed-citation>22. Blazquez MT, Anguianoa M, De Saavedra FA, Lallena AM, Carpenaet P. Study of the human postural control system during quiet standing using detrended fluctuation analysis. Physica A: statistical Mechanics and its Applications. 2009;388:1857-66.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Study of the human postural control system during quiet standing using detrended fluctuation analysis</article-title>
          <source>Physica A: statistical Mechanics and its Applications</source>
          <date>
            <year>2009</year>
          </date>
          <fpage>1857</fpage>
          <lpage>66</lpage>
          <volume>388</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Blazquez</surname>
              <given-names>MT</given-names>
            </name>
            <name>
              <surname>Anguianoa</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>De Saavedra</surname>
              <given-names>FA</given-names>
            </name>
            <name>
              <surname>Lallena</surname>
              <given-names>AM</given-names>
            </name>
            <name>
              <surname>Carpenaet</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B23">
        <mixed-citation>23. Tsuji H, Venditti FJ Jr, Manders ES, Evans JC, Larson MG, Feldman CL, et al. Determinants of heart rate variability. J Am Coll Cardiol. 1996;28(6):1539-46.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Determinants of heart rate variability</article-title>
          <source>J Am Coll Cardiol</source>
          <date>
            <year>1996</year>
          </date>
          <fpage>1539</fpage>
          <lpage>46</lpage>
          <issue>6</issue>
          <volume>28</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Tsuji</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Venditti Jr</surname>
              <given-names>FJ</given-names>
            </name>
            <name>
              <surname>Manders</surname>
              <given-names>ES</given-names>
            </name>
            <name>
              <surname>Evans</surname>
              <given-names>JC</given-names>
            </name>
            <name>
              <surname>Larson</surname>
              <given-names>MG</given-names>
            </name>
            <name>
              <surname>Feldman</surname>
              <given-names>CL</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B24">
        <mixed-citation>24. Thakre TP, Smith ML. Loss of lag-response curvilinearity of indices of heart rate variability in congestive heart failure. BMC Cardiovasc Disord. 2006;6:27.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Loss of lag-response curvilinearity of indices of heart rate variability in congestive heart failure</article-title>
          <source>BMC Cardiovasc Disord</source>
          <date>
            <year>2006</year>
          </date>
          <fpage>27</fpage>
          <volume>6</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Thakre</surname>
              <given-names>TP</given-names>
            </name>
            <name>
              <surname>Smith</surname>
              <given-names>ML</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B25">
        <mixed-citation>25. Lerma C, Infante O, Perez-Grovas H, Jose MV. Poincar&#233; plot indexes of heart rate variability capture dynamic adaptations after haemodialysis in chronic renal failure patients. Clin Physiol Funct Imaging. 2003;23(2):72-80.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Poincar&#233; plot indexes of heart rate variability capture dynamic adaptations after haemodialysis in chronic renal failure patients</article-title>
          <source>Clin Physiol Funct Imaging</source>
          <date>
            <year>2003</year>
          </date>
          <fpage>72</fpage>
          <lpage>80</lpage>
          <issue>2</issue>
          <volume>23</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Lerma</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Infante</surname>
              <given-names>O</given-names>
            </name>
            <name>
              <surname>Perez-Grovas</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Jose</surname>
              <given-names>MV</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B26">
        <mixed-citation>26. Brennan M, Palaniswami M, Kamen P. Poincar&#233; plot interpretation using a physiological model of HRV based on a network of oscillators. Am J Physiol Heart Circ Physiol. 2002;283(5):H1873-86.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Poincar&#233; plot interpretation using a physiological model of HRV based on a network of oscillators</article-title>
          <source>Am J Physiol Heart Circ Physiol</source>
          <date>
            <year>2002</year>
          </date>
          <fpage>H1873</fpage>
          <lpage>86</lpage>
          <issue>5</issue>
          <volume>283</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Brennan</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Palaniswami</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Kamen</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B27">
        <mixed-citation>27. Julius S, Nesbitt S. Sympathetic overactivity in hypertension: a moving target. Am J Hypertens. 1996;9(11):113S-120S.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Sympathetic overactivity in hypertension: a moving target</article-title>
          <source>Am J Hypertens</source>
          <date>
            <year>1996</year>
          </date>
          <fpage>113S</fpage>
          <lpage>120S</lpage>
          <issue>11</issue>
          <volume>9</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Julius</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Nesbitt</surname>
              <given-names>S</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B28">
        <mixed-citation>28. Triposkiadis F, Karayannis G, Giamouzis G, Skoularigis J, Louridas G, Butler J. The sympathetic nervous system in heart failure: physiology, pathophysiology, and clinical implications. J Am Coll Cardiol. 2009;54(19):1747-62.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>The sympathetic nervous system in heart failure: physiology, pathophysiology, and clinical implications</article-title>
          <source>J Am Coll Cardiol</source>
          <date>
            <year>2009</year>
          </date>
          <fpage>1747</fpage>
          <lpage>62</lpage>
          <issue>19</issue>
          <volume>54</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Triposkiadis</surname>
              <given-names>F</given-names>
            </name>
            <name>
              <surname>Karayannis</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Giamouzis</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Skoularigis</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Louridas</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Butler</surname>
              <given-names>J</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B29">
        <mixed-citation>29. Stefanovska A, Bracic M. Physics of the human cardiovascular system. Contemp Phys. 1999;40:31-55.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Physics of the human cardiovascular system</article-title>
          <source>Contemp Phys</source>
          <date>
            <year>1999</year>
          </date>
          <fpage>31</fpage>
          <lpage>55</lpage>
          <volume>40</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Stefanovska</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Bracic</surname>
              <given-names>M</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B30">
        <mixed-citation>30. Javorka M, Javorkova J, Tonhajzerova I, Calkovska A, Javorka K. Heart rate variability in young patients with diabetes mellitus and healthy subjects explored by Poincar&#233; and sequence plots. Clin Physiol Funct Imaging. 2005;25(2):119-27.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Heart rate variability in young patients with diabetes mellitus and healthy subjects explored by Poincar&#233; and sequence plots</article-title>
          <source>Clin Physiol Funct Imaging</source>
          <date>
            <year>2005</year>
          </date>
          <fpage>119</fpage>
          <lpage>27</lpage>
          <issue>2</issue>
          <volume>25</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Javorka</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Javorkova</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Tonhajzerova</surname>
              <given-names>I</given-names>
            </name>
            <name>
              <surname>Calkovska</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Javorka</surname>
              <given-names>K</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">Nonlinear methods to assess changes in heart rate variability in type 2 diabetic patients</article-title>
      </title-group>
      <abstract xml:lang="en">
        <p>BACKGROUND: Heart rate variability (HRV) is an important indicator of autonomic modulation of cardiovascular function. Diabetes can alter cardiac autonomic modulation by damaging afferent inputs, thereby increasing the risk of cardiovascular disease. We applied nonlinear analytical methods to identify parameters associated with HRV that are indicative of changes in autonomic modulation of heart function in diabetic patients. OBJECTIVE: We analyzed differences in HRV patterns between diabetic and age-matched healthy control subjects using nonlinear methods. METHODS: Methods: Lagged Poincar&#233; plot, autocorrelation, and detrended fluctuation analysis were applied to analyze HRV in electrocardiography (ECG) recordings. RESULTS: Lagged Poincar&#233; plot analysis revealed a decrease in the standard deviation of instantaneous beat-to-beat interval variability (SD1) and in the ratio of SD1 to the continuous long-term R-R interval variability (SD12) in the diabetic group, indicating a decrease in heart rate parasympathetic modulation. The detrended fluctuation exponent derived from long-term fitting was higher than the short-term one in the diabetic population, which was also consistent with decreased parasympathetic input. The autocorrelation function of the deviation of inter-beat intervals exhibited a highly correlated pattern in the diabetic group compared with the control group. CONCLUSIONS: The HRV pattern significantly differs between diabetic patients and healthy subjects. All three statistical methods employed in the study may prove useful to detect the onset and extent of autonomic neuropathy in diabetic patients.</p>
      </abstract>
      <kwd-group xml:lang="en">
        <kwd>Heart Failure</kwd>
        <kwd>Diabetes Mellitus</kwd>
        <kwd>Type 1</kwd>
        <kwd>Systole</kwd>
        <kwd>Measurements, Methods and Theories</kwd>
        <kwd>Statistics as Topic</kwd>
      </kwd-group>
    </front-stub>
    <body>
      <p>
        <fn id="enda-body2">
          <label>
            <bold>Nonlinear methods to assess changes in heart rate variability in type 2 diabetic patients</bold>
          </label>
          <p>
            <bold>Bhaskar Roy<sup>I,II</sup>; Sobhendu Ghatak<sup>I</sup></bold>
          </p>
        </fn>
      </p>
      <p><sup>I</sup>Indian Institute of Technology, &#205;ndia</p>
      <p><sup>II</sup>University of Connecticut, Farmington, CT, USA</p>
      <p>Mailing Address</p>
      <p>
        <bold>ABSTRACT</bold>
      </p>
      <p><bold>BACKGROUND:</bold> Heart rate variability (HRV) is an important indicator of autonomic modulation of cardiovascular function. Diabetes can alter cardiac autonomic modulation by damaging afferent inputs, thereby increasing the risk of cardiovascular disease. We applied nonlinear analytical methods to identify parameters associated with HRV that are indicative of changes in autonomic modulation of heart function in diabetic patients.</p>
      <p><bold>OBJECTIVE:</bold> We analyzed differences in HRV patterns between diabetic and age-matched healthy control subjects using nonlinear methods.</p>
      <p><bold>METHODS:</bold> Methods: Lagged Poincar&#233; plot, autocorrelation, and detrended fluctuation analysis were applied to analyze HRV in electrocardiography (ECG) recordings.</p>
      <p><bold>RESULTS:</bold> Lagged Poincar&#233; plot analysis revealed a decrease in the standard deviation of instantaneous beat-to-beat interval variability (SD1) and in the ratio of SD1 to the continuous long-term R-R interval variability (SD12) in the diabetic group, indicating a decrease in heart rate parasympathetic modulation. The detrended fluctuation exponent derived from long-term fitting was higher than the short-term one in the diabetic population, which was also consistent with decreased parasympathetic input. The autocorrelation function of the deviation of inter-beat intervals exhibited a highly correlated pattern in the diabetic group compared with the control group.</p>
      <p><bold>CONCLUSIONS:</bold> The HRV pattern significantly differs between diabetic patients and healthy subjects. All three statistical methods employed in the study may prove useful to detect the onset and extent of autonomic neuropathy in diabetic patients.</p>
      <p><bold>Keywords:</bold> Heart Failure; Diabetes Mellitus, Type 1; Systole; Measurements, Methods and Theories; Statistics as Topic.</p>
      <p>
        <bold>Introduction</bold>
      </p>
      <p>Heart rate is dynamically regulated by intrinsic and extrinsic control systems, maintaining homeostasis. The major extrinsic control is provided by the autonomic nervous system. Heart rate variability (HRV) is a measure of the fluctuation in the interval between sequential sinus heartbeats, and reflects cardiac autonomic regulation <sup>1-3</sup>. Diabetes leads to autonomic neuropathy <sup>4</sup>, thereby disrupting a major component of cardiovascular regulation and contributing to an increased incidence of cardiovascular diseases in diabetic patients, such as heart attack, sudden cardiac death, and silent ischemia <sup>5-8</sup>. Early diagnosis of autonomic diabetic neuropathy is difficult and the detection methods available, e.g., the Ewing Test Battery, are cumbersome and have poor sensitivity and reproducibility. In contrast, HRV analysis is noninvasive and the input data are easily obtained by conventional electrocardiography (ECG)<sup>9-12</sup>. However, because of the nonlinear heart dynamics, conventional time and frequency domain parameters of HRV may not always represent the nonstationary characteristics of ECG. Nonlinear methods such as the Poincar&#233; plot, detrended fluctuation analysis (DFA), tone/entropy analysis and HR complexity analysis are newly developed tools used for identifying nonlinear patterns within ECG data <sup>13-18</sup>.</p>
      <p>In this study, we used nonlinear analytical methods to study the differences in HRV patterns between diabetic and healthy individuals. The purpose of this study was to identify new parameters useful for detecting autonomic dysregulation in diabetes.</p>
      <p>
        <bold>Methods</bold>
      </p>
      <p>The patient group consisted of 23 type 2 diabetes mellitus patients with no history of cardiac, neurological, psychiatric, or sleep disorders. Patients on heart rate-altering medications were excluded from the study. The study was approved by the ethical committee of the Indian Institution of Technology, Kharagpur, India. A total of 23 healthy subjects were selected as a control group using the same exclusion criteria. All participants provided written informed consent prior to inclusion in the study. Subjects were instructed to avoid caffeine, alcohol, and physical exertion the day before the study was performed. A 10-min ECG recording was acquired from the patients while on supine position following a 15-min relaxation period. All ECGs were recorded at a fixed time of day to avoid the effects of diurnal variations on HRV. </p>
      <p>Matlab and SPSS software packages were used for statistical analysis. For comparative analysis between the groups, unpaired t-tests were applied as appropriate. Other statistical methods are individually described in details. </p>
      <p>
        <bold>Poincar&#233; Plot</bold>
      </p>
      <p>The Poincar&#233; plot is a scatter plot of <italic>RR</italic><sub>n</sub> vs. <italic>RR</italic><sub>n+1</sub> where <italic>RR</italic><sub>n</sub> is the time between two successive R peaks and <italic>RR</italic><sub>n+1</sub> is the time between the next two successive R peaks. When the plot is adjusted by the ellipse-fitting technique, the analysis provides three indices: the standard deviation of instantaneous beat-to-beat interval variability (SD1), the continuous long-term R/R interval variability (SD2), and the SD1/SD2 ratio (SD12)<sup>15</sup>. On the Poincar&#233; plot, SD1 it is the width and SD2 the length of the ellipse. In addition to this conventional plot (<italic>RR</italic><sub>n+1</sub> vs. <italic>RR</italic><sub>n</sub>), we also used the generalized Poincar&#233; plot with different intervals, including the m-lagged Poincar&#233; plot (the plot of <italic>RR</italic><sub>n+m</sub> versus <italic>RR</italic><sub>n</sub>). The values of SD1 and SD2 were calculated for lag = m from the relations </p>
      <p>SD1 = (&#934;(m) - &#934;(0))<sup>1/2</sup> and SD2 = (&#934;(m) + &#934;(0))<sup>1/2</sup>, where the autocovariance function &#934;(m) is given by</p>
      <p>
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e01-en.jpg"/></p>
      <p>and <inline-formula id="car01-body2"><inline-graphic xlink:href="https://minio.scielo.br/documentstore/1678-4170/zwb6ZdCrX9ZSDLBnb4fL58C/5dd9330853ebcfd958d3ea994acc6ca7e33bcaf7.jpg"/> is the mean RR</inline-formula><sub>n</sub><sup>14</sup>. For the purpose of our study, we set m at 1, 5, and 9. We then extended our analysis to reveal the association between these standard deviation (SD) values and m by using the Pad&#233; approximation<sup>19</sup>. We assumed a simple form of the Pad&#233; approximation for SD values as the ratio of polynomial in M of degree one.</p>
      <p>
        <disp-formula id="equ01-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e02-en.jpg"/></disp-formula>
      </p>
      <p>Here Y = SD1, SD2, or SD12 and &#967; = a/c. The terms &#946; = b/a and &#947; = d/c are the new unknown parameters. In order to determine if these parameters are of value for assessing cardiovascular health, we considered eq. (1) for the case of small m. In this limit, equation (1) can be approximated as Y = C + LM + QM<sup>2</sup>, where the slope is L = &#967; (&#946; - &#947;) and the curvature is Q = &#947;L. The slope and curvature of the plot of SD vs. m were determined by the fitted parameters &#967;, &#946;, and &#947;.</p>
      <p>
        <bold>Detrended Fluctuation Analysis</bold>
      </p>
      <p>Another analytic method to assess long-term correlation in the R-R-time sequence is based on DFA <sup>20</sup>. The measure of correlation was given by a scaling exponent (&#945;) of the fluctuation function F(&#964;) &#8776; &#964;<sup>&#945;</sup>. The fluctuation function F(&#964;) was computed as follows. For a given time sequence R(t<sub>i</sub>), t<sub>i</sub> = i&#948;t, where &#948;t is the characteristic time interval for the sequence and i = 1, N is an integrated time series, r(t<sub>i</sub>) was defined as r(t<sub>i</sub>) = &#931;<sup>i</sup><sub>j</sub> [R(t<sub>j</sub>) - &lt;R&gt;], i = 1,N, where &lt;R&gt; is the mean of R(t<sub>i</sub>). The integrated series was divided into segments of equal duration, &#964; = n &#948;t and a linear function used to fit the data within each segment. The fluctuation function F(&#964;) was calculated as the root mean square fluctuation relative to the linear trend and alpha was obtained by fitting the data to a power law function. It has been observed that an acceptable estimate of the scaling exponent alpha (from DFA) can be obtained from analysis of data sets with 256 samples or longer (equivalent to approximately 3.5 min of RR data at a heart rate of 70 beats/min). The analysis of RR data from an ECG recording period of 10 min was therefore expected to provide an adequate measure of the scaling exponent <sup>21</sup>. However, the alpha value obtained from this calculation may be under the mixed influence of both short-term scaling, reflecting parasympathetic control, and long-term scaling, reflecting sympathetic control, and thus may fail to fully distinguish parasympathetic and sympathetic influences. A separate analysis of both short- and long-term scaling is supposed to nullify the mutual effect and reveal the exact scaling variation <sup>22</sup>. Thus, we analyzed separate alpha values, short-term &#945;<sub>s</sub> and long-term &#945;<sub>l</sub>. For &#945;<sub>s</sub>, data from 25 beats were included, whereas for &#945;<sub>l</sub>, data from 30 to N/4 beats were included.</p>
      <p>
        <bold>Correlation between successive differences in RRn interval</bold>
      </p>
      <p>The coherence of the RR<sub>n</sub> interval can be assessed from the map of interval variation: </p>
      <p>
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e03-en.jpg"/></p>
      <p>where &lt;RR<sub>n</sub>&gt; is the mean interval. This plot is expected to show the correlation between the variability of three consecutive R-R intervals.</p>
      <p>Autocorrelation of fluctuation of RRn</p>
      <p> We explored the autocorrelation of the deviation of RR<sub>n</sub> from the mean &lt;RR<sub>n</sub>&gt;. The autocorrelation function C(m) of a particular subject was calculated from</p>
      <p>
        <graphic xlink:href="1678-4170-abc-101-04-0317-30181-e04-en.jpg"/></p>
      <p>where the deviation is &#8710;RR<sub>n </sub>= RR<sub>n </sub>- (RR<sub>n</sub>) and N is the total number of RR<sub>n</sub> intervals.</p>
      <p>
        <bold>Results</bold>
      </p>
      <p>The mean heart rate was 74.7 &#144;&#177; 6.1 beats/min in the diabetic group and 72.4 &#144;&#177; 6.7 beats/min in the healthy control group. Mean age in the diabetic group was 46.3 years (range, 36-56 years) and 47.4 years (range, 39-57 years) in the control group. All study subjects were normotensive.</p>
      <p>In the Poincar&#233; plot analysis, plot scatter increased with lag number, yielding higher width (SD1) and length (SD2) values. The incremental increase in width of the plot <italic>RR</italic><sub>n+m</sub> vs. <italic>RR</italic><sub>n</sub> as m increased was smaller in the diabetes group (<xref ref-type="fig" rid="aop_5014fig01-body2">Figure 1</xref>, D) than in the control group (<xref ref-type="fig" rid="aop_5014fig01-body2">Figure 1</xref>, ND). Differences in the values of SD1, SD2, and SD12 between the diabetes group and the control group were statistically significant (p &lt; 0.001 for all). The values of SD1 and SD12 were higher in the control group, whereas SD2 was higher in the diabetic group. The difference in SD12 increased with lag number (<xref ref-type="fig" rid="aop_5014fig02-body2">Figure 2</xref>).</p>
      <p content-type="asset">
        <fig id="aop_5014fig01-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf01-en.jpg"/></fig>
      </p>
      <p content-type="asset">
        <fig id="aop_5014fig02-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf02-en.jpg"/></fig>
      </p>
      <p>An excellent fit of the data with equation (1) (solid line on the curve, R<sup>2</sup> = 0.999) was found with the &#967;, &#946;, &#947; value sets listed in <xref ref-type="table" rid="en_aop_5014tab01-body2">Table 1</xref>. The values for L and Q as obtained by fitting of the data to eq. (1) are also presented in <xref ref-type="table" rid="en_aop_5014tab01-body2">Table 1</xref>. The general features were that the slope (L) was positive but curvature (Q) was negative for all parameters and curvature was nearly one order of magnitude smaller than the slope.</p>
      <p content-type="asset">
        <table-wrap id="en_aop_5014tab01-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gt01-en.jpg"/></table-wrap>
      </p>
      <p>From DFA, the mean value of alpha in the control group was smaller than that in the diabetic group (0.88 &#144;&#177; 0.17 vs. 1.02 &#144;&#177; 0.13; p &lt; 0.001) (<xref ref-type="fig" rid="aop_5014fig03-body2">Figure 3</xref>). In control subjects, &#945;<sub>s</sub> was slightly larger than &#945;<sub>l</sub> (1.01 &#144;&#177; 0.14 vs. 0.80 &#144;&#177; 0.19), whereas &#945;<sub>l</sub> was larger than &#945;<sub>s</sub> for the diabetic group (&#945;<sub>s</sub> = 1.09 &#144;&#177; 0.17; &#945;<sub>l</sub> = 1.18 &#144;&#177; 0.19). When &#945;<sub>s</sub> was plotted against &#945;<sub>l</sub> (<xref ref-type="fig" rid="aop_5014fig04-body2">Figure 4</xref>), the diabetic and nondiabetic populations tended to form two separate clusters.</p>
      <p content-type="asset">
        <fig id="aop_5014fig03-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf03-en.jpg"/></fig>
      </p>
      <p content-type="asset">
        <fig id="aop_5014fig04-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf04-en.jpg"/></fig>
      </p>
      <p>In the correlation plot, points were crowded around the origin for diabetic patients. In contrast, there was greater scattering about the origin and more asymmetry in the plot of control subjects (<xref ref-type="fig" rid="aop_5014fig05-body2">Figure 5</xref>, ND1, ND2). The strength of heart rhythm correlation was estimated by considering the autocorrelation of fluctuation in RR<sub>n</sub>. Representative results from one control and one diabetic patient are plotted in <xref ref-type="fig" rid="aop_5014fig06-body2">Figure 6</xref>. The autocorrelation functions for diabetic and control patients were distinct. For diabetic subjects, the correlation function C(m) decreased slowly (black and green curve in the upper figure) with lag time. The time dependence was close to the sum of the two exponentials with superimposed small amplitude oscillation of low frequency. On the other hand, C(m) from the healthy subjects demonstrated a more rapid (exponential) fall as correlation time decreased compared with the diabetic cases. To confirm this difference in correlation pattern between control and diabetic subjects, we shuffled the actual time series of R-R interval using Matlab software and the autocorrelation functions of the shuffled data (red and blue for subjects 1 and 2 respectively) were plotted in <xref ref-type="fig" rid="aop_5014fig06-body2">Figure 6</xref>. The autocorrelation functions of the shuffled data from all subjects (2 diabetics and 2 healthy controls) were nearly identical. </p>
      <p content-type="asset">
        <fig id="aop_5014fig05-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf05-en.jpg"/></fig>
      </p>
      <p content-type="asset">
        <fig id="aop_5014fig06-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf06-en.jpg"/></fig>
      </p>
      <p>We also characterized properties of &#8710;RRn by the probability distribution function P(&#8710;RRn) (<xref ref-type="fig" rid="aop_5014fig07-body2">Figure 7</xref>). For diabetic patients, the probability distribution was almost symmetrical and could be fit by a Gaussian function (R<sup>2</sup> = 0.93) with width = 0.023. For healthy subjects, the probability distribution P was asymmetrical with positive mean and higher width = 0.036 as obtained by the Gaussian fit (R<sup>2</sup> = 0.93).</p>
      <p content-type="asset">
        <fig id="aop_5014fig07-body2">
          <graphic xlink:href="1678-4170-abc-101-04-0317-30181-gf07-en.jpg"/></fig>
      </p>
      <p>
        <bold>Discussion</bold>
      </p>
      <p>We found marked differences in HRV pattern between diabetic and healthy control subjects using nonlinear analyses. Subjects were matched for both mean age and resting heart rate, the two major determinants of HRV <sup>23</sup>, so that the difference in distribution would reflect changes in cardiovascular regulation resulting from the diabetic condition only.</p>
      <p>Several modifications of the simple Poincar&#233; plot have been proposed to more effectively reveal changes in HRV patterns, including the lagged plot. The concept of this m lagged plot emerged from the recognition that any given R-R interval can influence up to eight subsequent R-R intervals <sup>24,25</sup>. It has been shown that SD1 correlates with the short-term variability of heart rate and is mainly influenced by parasympathetic modulation, whereas SD2 is a measure of long-term variability <sup>14,26</sup> and reflects sympathetic activation. The lower SD1 in diabetic subjects indicates that parasympathetic regulation is weakened by the disease, presumably by peripheral neuropathy, whereas higher SD2 in diabetic patients indicates increased long-term variability because of compensatory sympathetic input. </p>
      <p>The results from Poincar&#233; plot analysis are further revealed by the slope (L) and curvature (-Q) of the plot. In the diabetic group, L and -Q for SD1 and SD12 were smaller, whereas L and -Q values for SD2 were higher than in the control group. The difference in Q was larger than the difference in L. In particular, the Q value for SD12 in the control group was &gt;3 times greater than that for diabetic group. Low values of curvature are found in patients with cardiovascular disease <sup>24</sup>. These data strongly suggest decreased parasympathetic activity and excessive influence of sympathetic activity in the diabetic heart. In addition, this result provides indirect support for the notion that higher sympathetic influence over cardiovascular function is correlated with cardiac morbidity <sup>27,28</sup>. An increased SD12 is considered a good indicator of healthy heart dynamics, and the lower value in diabetic patients again supports altered sympathovagal balance in diabetes.</p>
      <p>Previous reports using DFA showed that &#945;<sub>s</sub> &gt; &#945;<sub>l</sub> in healthy subjects, whereas the reverse was the case for subjects with cardiovascular disease <sup>20</sup>. We found a similar trend in this study, again confirming the adverse effect of diabetes on the heart. </p>
      <p>In the absence of external modulation, the correlation plot is expected to scatter close to the point of origin, whereas random input will produce a uniform distribution. We observed a high density of points around the origin with greater symmetry in diabetic patients when compared with controls. Plots from healthy controls were generally asymmetrically scattered with large RR<sub>n</sub> values. These results suggest that mechanisms for decelerating and accelerating HR over different time frames are substantially impaired in diabetic patients.</p>
      <p>Application of autocorrelation to HRV analysis is a recent idea that regards HRV as the outcome of the interaction between coupled oscillators of various frequencies <sup>29</sup>. The degree of autocorrelation can also reflect on the embedded time scales within the HRV pattern. It is thought that each of these time scales in the coupled oscillator is represented by a separate self-oscillator, interacting with other oscillators with different physiological functions <sup>18</sup>. The lack of exponential fall in C(m) indicates the presence of a long-term memory effect in the diabetic condition and strongly suggests that mechanisms for short-term variation in heart rate are weakened or lacking in diabetic patients.</p>
      <p>Heart rate variability analysis based on nonlinear dynamics has been shown to be superior to conventional methods for identifying hidden changes in cardiac autonomic modulation in various disease conditions. Previous reports have demonstrated differences in Poincar&#233; plots, DFA, and sample entropy analysis between the hearts of diabetic and nondiabetic patients <sup>17,30</sup>, but these differences often did not reach statistical significance because of the small sample sizes <sup>17</sup>. Our study not only enrolled larger numbers of patients and controls but also used multiple nonlinear analytic tools, including Poincar&#233; plot analysis, DFA, and autocorrelation analysis to reveal changes in HRV due to diabetic neuropathy.</p>
      <p>The major limitation of this study is the heterogeneous patient population. The duration of illness in the patient group was variable and many were on different antidiabetic medications. Moreover, a population of 23 patients may be sufficient to identify differences in HRV pattern between diabetic and healthy nondiabetic subjects, but a much larger group of patients is required to confirm the true diagnostic and prognostic values of the parameters derived from the analytic methods. Intra-group analysis in a larger group of diabetic patients of variable disease duration to assess progressive changes in HRV pattern is the next logical step. Our study establishes the potential of nonlinear methods of heart rate variability analysis to assess changes in HRV pattern indicative of cardiovascular disease, including effects associated with diabetes mellitus.</p>
      <p>
        <bold>Conclusions</bold>
      </p>
      <p>In summary, we have shown the effectiveness of nonlinear analytical methods to study differences in HRV patterns between diabetic patients and healthy-matched controls. We also emphasized the novelty of autocorrelation analysis to assess changes in the autonomic regulation of the diabetic heart. To our knowledge, this is the first attempt to distinguish normal from diabetic heart function using autocorrelation analysis. We believe these methods have the potential to identify diagnostic and prognostic markers for cardiac autonomic neuropathy in diabetes.</p>
      <p>
        <bold>Author contributions</bold>
      </p>
      <p>Conception and design of the research, Acquisition of data, Analysis and interpretation of the data, Statistical analysis, Writing of the manuscript, Critical revision of the manuscript for intellectual content: Roy, B, Ghatak S.</p>
      <p>
        <bold>Potential Conflict of Interest</bold>
      </p>
      <p>No potential conflict of interest relevant to this article was reported.</p>
      <p>
        <bold>Sources of Funding</bold>
      </p>
      <p>There were no external funding sources for this study.</p>
      <p>
        <bold>Study Association</bold>
      </p>
      <p>This study is not associated with any post-graduation program.</p>
      <p>
        <bold>References</bold>
      </p>
    </body>
  </sub-article>
</article>
