<!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:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" article-type="research-article" dtd-version="1.0" specific-use="sps-1.2" xml:lang="en">
  <front>
    <journal-meta>
      <journal-id journal-id-type="nlm-ta">Arq Bras Cardiol</journal-id>
      <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</publisher-name>
      </publisher>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v3">xgNZY538QwjDk6Jyg9DgdNx</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v2">S0066-782X2015002500552</article-id>
      <article-id pub-id-type="publisher-id" specific-use="previous-pid">S0066-782X2015005050108</article-id>
      <article-id pub-id-type="doi">10.5935/abc.20150108</article-id>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Original Articles</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>Echocardiographic Predictors of Worse Outcome After Cardiac Resynchronization Therapy</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <name>
            <surname>Rocha</surname>
            <given-names>Eduardo Arrais</given-names>
          </name>
          <xref ref-type="aff" rid="aff01">1</xref>
          <xref ref-type="aff" rid="aff02">2</xref>
          <xref ref-type="corresp" rid="c01"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Pereira</surname>
            <given-names>Francisca Tatiana Moreira</given-names>
          </name>
          <xref ref-type="aff" rid="aff02">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Abreu</surname>
            <given-names>Jos&#233; Sebasti&#227;o</given-names>
          </name>
          <xref ref-type="aff" rid="aff02">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Lima</surname>
            <given-names>Jos&#233; Wellington O.</given-names>
          </name>
          <xref ref-type="aff" rid="aff03">3</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Monteiro</surname>
            <given-names>Marcelo de Paula M.</given-names>
          </name>
          <xref ref-type="aff" rid="aff02">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Rocha</surname>
            <given-names>Almino Cavalcante</given-names>
            <suffix>Neto</suffix>
          </name>
          <xref ref-type="aff" rid="aff02">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Quidute</surname>
            <given-names>Ana Rosa Pinto</given-names>
          </name>
          <xref ref-type="aff" rid="aff02">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Go&#233;s</surname>
            <given-names>Camilla Viana A.</given-names>
          </name>
          <xref ref-type="aff" rid="aff01">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Rodrigues</surname>
            <given-names>Carlos Roberto Martins</given-names>
            <suffix>Sobrinho</suffix>
          </name>
          <xref ref-type="aff" rid="aff02">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Scanavacca</surname>
            <given-names>Maur&#237;cio Ibrahim</given-names>
          </name>
          <xref ref-type="aff" rid="aff01">1</xref>
        </contrib>
      </contrib-group>
      <aff id="aff01">
        <label>1</label>
        <institution content-type="orgname">Universidade de S&#227;o Paulo</institution>
        <addr-line>
          <named-content content-type="city">S&#227;o Paulo</named-content>
          <named-content content-type="state">SP</named-content>
        </addr-line>
        <country country="BR">Brazil</country>
        <institution content-type="original">Universidade de S&#227;o Paulo &#8211; USP, S&#227;o Paulo, SP</institution>
        <institution content-type="normalized">Universidade de S&#227;o Paulo</institution>
      </aff>
      <aff id="aff02">
        <label>2</label>
        <institution content-type="orgname">Universidade Federal do Cear&#225;</institution>
        <addr-line>
          <named-content content-type="city">Fortaleza</named-content>
          <named-content content-type="state">CE</named-content>
        </addr-line>
        <country country="BR">Brazil</country>
        <institution content-type="original">Universidade Federal do Cear&#225; &#8211; UFC, Fortaleza, CE &#8211; Brazil</institution>
        <institution content-type="normalized">Universidade Federal do Cear&#225;</institution>
      </aff>
      <aff id="aff03">
        <label>3</label>
        <institution content-type="orgname">Universidade Estadual do Cear&#225;</institution>
        <addr-line>
          <named-content content-type="city">Fortaleza</named-content>
          <named-content content-type="state">CE</named-content>
        </addr-line>
        <country country="BR">Brazil</country>
        <institution content-type="original">Universidade Estadual do Cear&#225; &#8211; UECE, Fortaleza, CE &#8211; Brazil</institution>
        <institution content-type="normalized">Universidade Estadual do Cear&#225;</institution>
      </aff>
      <author-notes>
        <corresp id="c01"><bold>Mailing Address: Eduardo Arrais Rocha</bold> , Universidade Federal do Cear&#225;. Av. Padre Ant&#244;nio Tom&#225;s, 3535 / 1301, Coc&#243;. Postal Code 60192-120, Fortaleza, CE &#8211; Brazil. E-mail: <email>eduardoa@cardiol.br</email>, <email>eduardoarraisrocha@gmail.com</email>
					</corresp>
        <fn fn-type="con">
          <p>
            <bold>Author contributions</bold>
          </p>
          <p>Conception and design of the research: Rocha EA, Pereira FTM, Abreu JS, Lima JWO, Rocha Neto AC, Rodrigues Sobrinho CRM, Scanavacca MI; Acquisition of data: Rocha EA, Pereira FTM, Abreu JS, Monteiro MPM, Go&#233;s CVA; Analysis and interpretation of the data: Rocha EA, Pereira FTM, Abreu JS, Lima JWO, Monteiro MPM, Rocha Neto AC; Statistical analysis: Rocha EA, Lima JWO, Quidute ARP; Obtaining financing: Rocha EA; Writing of the manuscript: Rocha EA, Abreu JS, Rocha Neto AC, Quidute ARP, Rodrigues Sobrinho CRM, Scanavacca MI.</p>
        </fn>
        <fn fn-type="conflict">
          <p>
            <bold>Potential Conflict of Interest</bold>
          </p>
          <p>No potential conflict of interest relevant to this article was reported.</p>
        </fn>
      </author-notes>
      <pub-date pub-type="epub">
        <day>04</day>
        <month>09</month>
        <year>2015</year>
      </pub-date>
      <pub-date pub-type="collection">
        <month>12</month>
        <year>2015</year>
      </pub-date>
      <volume>105</volume>
      <issue>6</issue>
      <fpage>552</fpage>
      <lpage>559</lpage>
      <history>
        <date date-type="received">
          <day>11</day>
          <month>05</month>
          <year>2015</year>
        </date>
        <date date-type="rev-recd">
          <day>01</day>
          <month>07</month>
          <year>2015</year>
        </date>
        <date date-type="accepted">
          <day>01</day>
          <month>07</month>
          <year>2015</year>
        </date>
      </history>
      <permissions>
        <license license-type="open-access" xlink:href="http://creativecommons.org/licenses/by/4.0/">
          <license-p>This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.</license-p>
        </license>
      </permissions>
      <abstract>
        <title>Abstract</title>
        <sec>
          <title>Background:</title>
          <p>Cardiac resynchronization therapy (CRT) is the recommended treatment by leading global guidelines. However, 30%-40% of selected patients are non-responders.</p>
        </sec>
        <sec>
          <title>Objective:</title>
          <p>To develop an echocardiographic model to predict cardiac death or transplantation (Tx) 1 year after CRT.</p>
        </sec>
        <sec>
          <title>Method:</title>
          <p>Observational, prospective study, with the inclusion of 116 patients, aged 64.89 &#177; 11.18 years, 69.8% male, 68,1% in NYHA FC III and 31,9% in FC IV, 71.55% with left bundle-branch block, and median ejection fraction (EF) of 29%. Evaluations were made in the pre&#8209;implantation period and 6-12 months after that, and correlated with cardiac mortality/Tx at the end of follow-up. Cox and logistic regression analyses were performed with ROC and Kaplan-Meier curves. The model was internally validated by bootstrapping.</p>
        </sec>
        <sec>
          <title>Results:</title>
          <p>There were 29 (25%) deaths/Tx during follow-up of 34.09 &#177; 17.9 months. Cardiac mortality/Tx was 16.3%. In the multivariate Cox model, EF &lt; 30%, grade III/IV diastolic dysfunction and grade III mitral regurgitation at 6&#8209;12 months were independently related to increased cardiac mortality or Tx, with hazard ratios of 3.1, 4.63 and 7.11, respectively. The area under the ROC curve was 0.78.</p>
        </sec>
        <sec>
          <title>Conclusion:</title>
          <p>EF lower than 30%, severe diastolic dysfunction and severe mitral regurgitation indicate poor prognosis 1 year after CRT. The combination of two of those variables indicate the need for other treatment options.</p>
        </sec>
      </abstract>
      <kwd-group xml:lang="en">
        <title>Keywords:</title>
        <kwd>Heart Failure/ mortality</kwd>
        <kwd>Echocardiography</kwd>
        <kwd>Pacemaker, Artificial</kwd>
        <kwd>Cardiac Resynchronization Therapy</kwd>
        <kwd>Risk Factors</kwd>
      </kwd-group>
      <counts>
        <fig-count count="8"/>
        <table-count count="8"/>
        <equation-count count="0"/>
        <ref-count count="28"/>
        <page-count count="0"/>
      </counts>
    </article-meta>
  </front>
  <body>
    <sec sec-type="intro">
      <title>Introduction</title>
      <p>Scientific evidence places cardiac resynchronization therapy (CRT) as class I in the major guidelines of pacemaking and congestive heart failure (CHF) for patients on optimal clinical treatment, with <italic>New</italic> York Heart Association (NYHA) functional class (FC) II, III or ambulatory IV CHF, ejection fraction (EF) &#8804;35%, and intraventricular conduction disorder, mainly of the left bundle branch<xref ref-type="bibr" rid="B01">1</xref><sup>,</sup><xref ref-type="bibr" rid="B02">2</xref>.</p>
      <p>However, 30%-40% of the patients submitted to CRT might not have a favorable outcome, requiring surgical treatment, with high risks and costs, and no clinical, hemodynamic or survival benefits<xref ref-type="bibr" rid="B03">3</xref><sup>,</sup><xref ref-type="bibr" rid="B04">4</xref>.</p>
      <p>Several studies have been aimed at identifying predictors of response to CRT, but with different patterns of definition of response, using mainly comparisons with the end-systolic volume (ESV) and FC improvement<xref ref-type="bibr" rid="B05">5</xref>. However, patients considered to be responders according to the analysis of ventricular volumes have no correlation with the clinical improvement observed, the walking test or the quality of life. That discrepancy is not a surprise, considering that CRT acts via several hemodynamic and neuro-humoral pathways<xref ref-type="bibr" rid="B06">6</xref>.</p>
      <p>The effects of CRT involve changes in the systolic or diastolic mitral regurgitation grade, diastolic dysfunction grade, systolic function and dyssynchrony. Therefore, a prognostic assessment or response definition using only one variable, such as left ventricular ESV, lacks accuracy.</p>
      <p>Thus, multifactorial indices or scores<xref ref-type="bibr" rid="B07">7</xref> are required to more accurately identify predictors of better survival and to designate real responders. Those indices should involve a higher number of variables, with greater sensitivity and specificity.</p>
      <p>This study aimed at developing an echocardiographic model to predict the risk of cardiac death or transplantation (Tx) after CRT.</p>
    </sec>
    <sec sec-type="methods">
      <title>Methods</title>
      <p>This is a prospective, observational study, assessing 116 patients with multisite pacemaker implanted at a university-affiliated tertiary hospital. Assessments were performed in the pre-implantation period (1<sup>st</sup> analysis) and at 1 year after implantation (2<sup>nd</sup> analysis). <xref ref-type="table" rid="t01">Table 1</xref>shows the characteristics of the study population.</p>
      <table-wrap id="t01">
        <label>Table 1</label>
        <caption>
          <title>Baseline characteristics of the study population</title>
        </caption>
        <table frame="hsides" rules="groups">
          <thead>
            <tr>
              <th align="left">
                <bold>Baseline Variables</bold>
              </th>
              <th align="center">
                <bold>Results</bold>
              </th>
            </tr>
          </thead>
          <tbody>
            <tr>
              <td>Total of patients</td>
              <td align="center">116</td>
            </tr>
            <tr>
              <td>Male sex</td>
              <td align="center">69.83%</td>
            </tr>
            <tr>
              <td>Age</td>
              <td align="center">64.8 &#177; 11.1</td>
            </tr>
            <tr>
              <td>FC (NYHA) III</td>
              <td align="center">68.1%</td>
            </tr>
            <tr>
              <td>FC (NYHA) IV</td>
              <td align="center">31.9%</td>
            </tr>
            <tr>
              <td>Beta-blocker use</td>
              <td align="center">88.7%</td>
            </tr>
            <tr>
              <td>ACEI use</td>
              <td align="center">97.4%</td>
            </tr>
            <tr>
              <td>High doses of diuretics</td>
              <td align="center">31.9%</td>
            </tr>
            <tr>
              <td>Number of previous hospitalizations</td>
              <td align="center">108</td>
            </tr>
            <tr>
              <td>Chagas heart disease</td>
              <td align="center">11.2%</td>
            </tr>
            <tr>
              <td>Ischemic cardiomyopathy</td>
              <td align="center">29.3%</td>
            </tr>
            <tr>
              <td>Dilated cardiomyopathy</td>
              <td align="center">59.4%</td>
            </tr>
            <tr>
              <td>Previous QRS width</td>
              <td align="center">160 ms</td>
            </tr>
            <tr>
              <td>LBBB</td>
              <td align="center">71.50%</td>
            </tr>
            <tr>
              <td>EF (Simpson)</td>
              <td align="center">29%</td>
            </tr>
            <tr>
              <td>Diastolic dysfunction (grade III/ IV)</td>
              <td align="center">41.5%</td>
            </tr>
            <tr>
              <td>MR (grade II and III)</td>
              <td align="center">46%</td>
            </tr>
            <tr>
              <td>LVDD</td>
              <td align="center">70 mm</td>
            </tr>
            <tr>
              <td>Systolic BP</td>
              <td align="center">115 &#177; 17 mmHg</td>
            </tr>
            <tr>
              <td>Posterolateral vein</td>
              <td align="center">45.4%</td>
            </tr>
            <tr>
              <td>Creatinine</td>
              <td align="center">1.1 mg/dL</td>
            </tr>
          </tbody>
        </table>
        <table-wrap-foot>
          <fn id="TFN01">
            <p>FC: New York Heart Association (NYHA) functional class; ACEI: Angiotensin-converting enzyme inhibitor; LBBB: Left bundle-branch block; EF: Ejection fraction; MR: Mitral regurgitation; LVDD: Left ventricular diastolic diameter; BP: Blood pressure. Pre- and post-CRT QRS width, ejection fraction and LVDD were expressed as medians (non-normal variables). Age was presented as mean and standard deviation.</p>
          </fn>
        </table-wrap-foot>
      </table-wrap>
      <p>Most right ventricular electrodes were positioned in the apical region (84%), and a larger distance was kept between the left ventricular and right ventricular electrodes. The models used were St. Jude Medical, Biotronik, Medtronic, and Guidant in 92, 12, 10, and 2 patients, respectively. Two patients received implantations via mini-thoracotomy. The number of patients selected was based on previous studies with similar outcomes.</p>
      <p>Drug doses were modified during the follow-up at the discretion of the attending physicians (3 specialists on cardiac stimulation belonging to the same team), responsible for all clinical and surgical procedures. Six echocardiographic variables considered to be of easy acquisition in daily practice and of clinical usefulness were selected for analysis.</p>
      <sec>
        <title>Echocardiographic Parameters</title>
        <p>The echocardiographic parameters were analyzed according to the Brazilian<xref ref-type="bibr" rid="B08">8</xref> and North-American guidelines for echocardiography<xref ref-type="bibr" rid="B09">9</xref>. All echocardiographies were performed by 3 experienced echocardiographers, one of whom was responsible for 70% of the examinations. The GE Vivid 7 Ultrasound System (GE Healthcare, Fairfield, CT, USA) with a 3.5-MHz transducer was used. The echocardiographers, who were blinded to the previous clinical and echocardiographic findings of the patients, had experience with that type of patient.</p>
        <p>The systolic function was analyzed by using the Simpson method, in the two- and four-chamber two-dimensional mode, followed by the mean. The ventricular diameters were obtained in M mode, according to the guideline recommendations<xref ref-type="bibr" rid="B09">9</xref>.</p>
        <p>The diastolic function was assessed by using mitral flow (at rest and after Valsalva maneuver), tissue Doppler and flow propagation speed on color M-mode. The diastolic dysfunction was classified into four grades as follows: I, mild; II, moderate or pseudonormal; III, accentuated or with restrictive dysfunction; and IV, severe or with irreversible restrictive dysfunction<xref ref-type="bibr" rid="B10">10</xref>.</p>
        <p>The grade of mitral regurgitation was assessed by use of color Doppler, according to the left atrial filling percentage. In mild reflow, that percentage was below 20%; in moderate, between 20% and 40%; percentages greater than those indicated significant reflow. In that practical context, the Coanda effect was interpreted as moderate reflow when restricted to the atrial lateral wall, and accentuated, when extended to the upper pole of the left atrium.</p>
        <p>All patients provided written informed consent, and the study was approved by the Ethics Committee of the hospital.</p>
      </sec>
      <sec>
        <title>Statistical Analysis</title>
        <p>The Shapiro-Wilk test was used to check the normality of the distribution of the variables. Ejection fraction and left ventricular diastolic diameter (LVDD) did not have a normal distribution.</p>
        <p>The categorical variables were presented as frequencies and percentages, and the continuous variables, as means and standard deviations or medians. The categorical variables were compared by using the Mac-Nemar, Stuart-Maxwell or chi-square tests. The Student <italic>t</italic> test was used to compare the distributions of normal, continuous variables, while the Wilcoxon/Mann-Whitney test, to compare continuous variables without normal distribution. The distributions were considered significantly different for p values lower than 0.05.</p>
        <p>The univariate relationship between the echocardiographic variables and cardiac death or Tx was assessed by use of Kaplan-Meier survival curve, log-rank test and Cox regression. The continuous variables were assessed by use of Cox regression in the search for a cutoff point.</p>
        <p>A Cox multiple regression model was developed in the first year after CRT to assess the independent contribution of each significant echocardiographic variable previously selected in the Cox univariate model. Variables with a p value &lt; 10% were considered potential confounders. Each variable was included in the multivariate model according to hazard descending order, being excluded when p &#8805; 5%.</p>
        <p>Logistic regression analysis was performed, using hazard<xref ref-type="bibr" rid="B11">11</xref> as an independent variable to measure the risk, and cardiac death/Tx as a dependent variable. The accuracy of the models was assessed by using the ROC curve, with sensitivity and specificity. The model was elaborated by dividing the hazard scores into risk categories according to the number of existing variables and classified as low (class A), intermediate (class B) and high risk (class C). To elaborate the scores, the hazard of individual variables was divided by the highest hazard of the model, multiplied by 100 and rounded to the nearest highest algorithm.</p>
        <p>To assess the proportional hazards associated with the predictors, Schoenfeld test was used. Bootstraping was used for the internal validation of the model, with confidence interval for the estimated hazard, generating 10,000 random samples of size 60, without replacing the original data set. For each 10,000 artificial samples, the hazards corresponding to each covariable were estimated. The values were ordinated for each covariable, the 95% confidence interval being observed.</p>
        <p>Data were assessed by using the Stata/SE software, version 12.1 (StataCorp LP, College Station, TX, USA), and the &#8220;R&#8221; software (<italic>R Foundation for Statistical Computing</italic>, Vienna, Austria).</p>
      </sec>
    </sec>
    <sec sec-type="results">
      <title>Results</title>
      <p>In the 34.09 &#177; 17.9 months of follow-up, 29 patients died, accounting for a total mortality of 25%. Cardiac mortality/Tx was 16.3%, corresponding to 19 patients, 6 of whom underwent Tx during the study period, 5 due to refractory CHF and 1 due to arrhythmic storm. Of the 6 transplanted patients, 3 died due to advanced disease when undergoing Tx.</p>
      <p>In the Cox multivariate model, the variables EF &lt; 30%, diastolic dysfunction and mitral regurgitation were independently related to increased cardiac mortality/Tx, with hazard ratios of 3.1, 4.63 and 7.11, respectively (<xref ref-type="table" rid="t02">Table 2</xref>).</p>
      <table-wrap id="t02">
        <label>Table 2</label>
        <caption>
          <title>Model with the echocardiographic variables in the first year</title>
        </caption>
        <table frame="hsides" rules="groups">
          <thead>
            <tr>
              <th align="left">
                <bold>Variable</bold>
              </th>
              <th align="center">
                <bold>Hazard</bold>
              </th>
              <th align="center">
                <bold>p value</bold>
              </th>
              <th align="center">
                <bold>95% CI</bold>
              </th>
            </tr>
          </thead>
          <tbody>
            <tr>
              <td>MR</td>
              <td align="center">7.115132</td>
              <td align="center">0.001</td>
              <td align="center">2.26449 - 22.35604</td>
            </tr>
            <tr>
              <td>Diastolic dysfunction</td>
              <td align="center">4.631782</td>
              <td align="center">0.004</td>
              <td align="center">1.631656 - 13.14824</td>
            </tr>
            <tr>
              <td>EF &lt; 30%</td>
              <td align="center">3.101647</td>
              <td align="center">0.035</td>
              <td align="center">1.083580 - 8.878182</td>
            </tr>
          </tbody>
        </table>
        <table-wrap-foot>
          <fn id="TFN02">
            <p>CI: Confidence interval; Hazard: Hazard ratio; MR: Mitral regurgitation grade III (severe) as compared to grades II and I (moderate to mild); diastolic dysfunction - grades III and IV (severe) as compared to grades I and II (mild to moderate dysfunction); EF: Ejection fraction.</p>
          </fn>
        </table-wrap-foot>
      </table-wrap>
      <p>The significant variables in the multivariate model were also significant separately in the Kaplan-Meier model, when compared by using log-rank test (p &lt; 0.001) (<xref ref-type="fig" rid="f01">Figures 1</xref>-<xref ref-type="fig" rid="f03">3</xref>). The analysis of the model by using the ROC curve showed an area under the curve (AUC) of 0.785, with sensitivity of 56.2%, specificity of 94.1% and accuracy of 88.2%.</p>
      <fig id="f01">
        <label>Figure 1</label>
        <caption>
          <title>Kaplan-Meier curve for the variable ejection fraction (EF) dichotomized between &gt; 30% and &#8804;30% at the 2nd phase of the analysis (1 year) exclusively with the echocardiographic variables, p &lt; 0.001, compared by using log-rank test.</title>
        </caption>
        <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf01.jpg"/></fig>
      <fig id="f02">
        <label>Figure 2</label>
        <caption>
          <title>Kaplan-Meier curve for the variable diastolic dysfunction (grades I and II compared to grades III and IV) at the 2nd phase of the analysis (1 year), with p &lt; 0.001, compared by using log-rank test.</title>
        </caption>
        <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf02.jpg"/></fig>
      <fig id="f03">
        <label>Figure 3</label>
        <caption>
          <title>Kaplan-Meier curve for the variable mitral regurgitation (MR), comparing the grades mild, moderate and severe at the 2nd phase of the analysis (1 year), with p &gt; 0.001, by using log-rank test.</title>
        </caption>
        <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf03.jpg"/></fig>
      <p>For the models proposed, all variables were assessed for compliance with the proportional hazards assumption by using the Schoenfeld test (<xref ref-type="table" rid="t03">Table 3</xref>), with results confirming the model adjustment for the variables proposed. The 95% confidence interval was obtained and confirmed the statistical significance of the estimates of the proportions. Thus, the model was validated by bootstraping and showed no lack of adjustment or exaggerated sensitivity of data.</p>
      <table-wrap id="t03">
        <label>Table 3</label>
        <caption>
          <title>Proportional hazards test with the echocardiographic variables</title>
        </caption>
        <table frame="hsides" rules="groups">
          <thead>
            <tr>
              <th align="left">
                <bold>Variable</bold>
              </th>
              <th align="center">
                <bold>H</bold>
              </th>
              <th align="center">
                <bold>chi<sup>2</sup></bold>
              </th>
              <th align="center"/>
              <th align="center"/>
              <th align="center">
                <bold>df</bold>
              </th>
              <th align="center"/>
              <th align="center"/>
              <th align="center">
                <bold>p value</bold>
              </th>
            </tr>
          </thead>
          <tbody>
            <tr>
              <td>MR</td>
              <td align="center">-0.01674</td>
              <td align="center">0.00</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">1</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">0.9456</td>
            </tr>
            <tr>
              <td>Diastolic dysfunction</td>
              <td align="center">-0.43109</td>
              <td align="center">2.52</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">1</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">0.1127</td>
            </tr>
            <tr>
              <td>EF &lt; 30%</td>
              <td align="center">-0.09445</td>
              <td align="center">0.14</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">1</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">0.7096</td>
            </tr>
            <tr>
              <td>Result</td>
              <td align="center"/>
              <td align="center">2.84</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">3</td>
              <td align="center"/>
              <td align="center"/>
              <td align="center">0.4174</td>
            </tr>
          </tbody>
        </table>
        <table-wrap-foot>
          <fn id="TFN03">
            <p>Shoenfeld test; chi<sup>2</sup>: Chi-square test; p value: Statistical significance level; H: Baseline hazard; df: Degree of freedom; MR: Mitral regurgitation; EF: Ejection fraction.</p>
          </fn>
        </table-wrap-foot>
      </table-wrap>
      <p>From the combination of those variables, we elaborated a model and score with 3 classes as follows: class A (low risk for cardiac death/Tx), corresponding to the absence of the significant variables from multivariate analysis, implying an event-free rate (EFR) of 97.5% in 30 months. The presence of 1 variable (class B) implied an EFR of 83.1% in 30 months, and the combination of 2 or 3 variables (class C) implied an EFR of 38.5% in 30 months (<xref ref-type="fig" rid="f04">Figure 4</xref> and <xref ref-type="table" rid="t04">Table 4</xref>).</p>
      <fig id="f04">
        <label>Figure 4</label>
        <caption>
          <title>Model with the echocardiographic variables at the first year: Class A &#8211; no variable (low risk of cardiac death or transplantation); Class B &#8211; presence of 1 variable (intermediate risk); class C (high risk) &#8211; presence of 2 or 3 variables (ejection fraction &lt; 30%, grades III and IV diastolic dysfunction compared to grades I and II, and grade III mitral regurgitation compared to grades II and I). Class A implies an event-free rate (EFR) of 97.5% in 30 months, Class B implies an EFR of 83.1%, and Class C, an EFR of 38.5% in 30 months.</title>
        </caption>
        <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf04.jpg"/></fig>
      <table-wrap id="t04">
        <label>Table 4</label>
        <caption>
          <title>Score with the echocardiographic variables in the first year</title>
        </caption>
        <table frame="hsides" rules="groups">
          <thead>
            <tr>
              <th align="left">
                <bold>Variable</bold>
              </th>
              <th align="center">
                <bold>Hazard</bold>
              </th>
              <th align="center">
                <bold>N</bold>
              </th>
              <th align="center">
                <bold>Score</bold>
              </th>
              <th align="center">
                <bold>Class</bold>
              </th>
              <th align="center">
                <bold>Risk</bold>
              </th>
            </tr>
          </thead>
          <tbody>
            <tr>
              <td>None</td>
              <td align="center">1.0</td>
              <td align="center">62</td>
              <td align="center">0</td>
              <td align="center">A</td>
              <td align="center">Low</td>
            </tr>
            <tr>
              <td>EF &lt; 30 %</td>
              <td align="center">3.1</td>
              <td align="center">20</td>
              <td align="center">3</td>
              <td align="center">B</td>
              <td align="center">Intermediate</td>
            </tr>
            <tr>
              <td>DD</td>
              <td align="center">4.6</td>
              <td align="center">3</td>
              <td align="center">5</td>
              <td align="center">B</td>
              <td align="center">Intermediate</td>
            </tr>
            <tr>
              <td>MR</td>
              <td align="center">7.1</td>
              <td align="center">3</td>
              <td align="center">7</td>
              <td align="center">B</td>
              <td align="center">Intermediate</td>
            </tr>
            <tr>
              <td>EF + DD</td>
              <td align="center">14.3</td>
              <td align="center">7</td>
              <td align="center">8</td>
              <td align="center">C</td>
              <td align="center">High</td>
            </tr>
            <tr>
              <td>EF + MR</td>
              <td align="center">22.0</td>
              <td align="center">4</td>
              <td align="center">10</td>
              <td align="center">C</td>
              <td align="center">High</td>
            </tr>
            <tr>
              <td>DD + MR</td>
              <td align="center">32.9</td>
              <td align="center">1</td>
              <td align="center">12</td>
              <td align="center">C</td>
              <td align="center">High</td>
            </tr>
            <tr>
              <td>EF + MR + DD</td>
              <td align="center">102.2</td>
              <td align="center">2</td>
              <td align="center">15</td>
              <td align="center">C</td>
              <td align="center">High</td>
            </tr>
          </tbody>
        </table>
        <table-wrap-foot>
          <fn id="TFN04">
            <p>Hazard: Proportional hazards; EF: Ejection fraction; DD: Diastolic dysfunction grades III and IV (severe) compared to grades I and II (mild to moderate); MR: Mitral regurgitation grade III (severe) compared to grades II and I (moderate to mild); Class A - low risk of cardiac death or transplantation; Class B - intermediate risk of cardiac death or transplantation; Class C - high risk of cardiac death or transplantation. Hazard was used as an independent variable in the logistic regression model to elaborate the score. The score was obtained by dividing the variable &#8216;proportional hazard&#8217; by the highest value, multiplying by 100 and rounding to the nearest number The scores of the combined variables resulted from adding their individual values.</p>
          </fn>
        </table-wrap-foot>
      </table-wrap>
    </sec>
    <sec sec-type="discussion">
      <title>Discussion</title>
      <p>Three important echocardiographic variables (grade III mitral regurgitation<xref ref-type="bibr" rid="B12">12</xref>, grade III/IV diastolic dysfunction<xref ref-type="bibr" rid="B13">13</xref>, and EF &#8804;30%), when present 1 year after CRT, were predictors of cardiac death or Tx in this echocardiographic model.</p>
      <p>The model was developed in a high-risk population, which had a median EF of 29%, LVDD of 70 mm, severe diastolic dysfunction in 42% of the sample, moderate or severe mitral regurgitation in 45.6%, and hospitalization due to CHF in the past year in 64%.</p>
      <p>The overall mortality rate in this study was 25% (29/116) in 34 &#177; 17 months, while, Yu et al<xref ref-type="bibr" rid="B05">5</xref>, in a study including patients with EF of 40% and FC II, therefore at lower risk, have reported an overall mortality rate of 15.6% in 24 months. The CARE-HF study, with a follow-up of 29.4 months, has reported a mortality rate of 30% in the group without intervention as compared to 20% in the group undergoing CRT<xref ref-type="bibr" rid="B14">14</xref>. The COMPANION study, in a 24-month follow-up, has reported a mortality rate of 21% (131/617) in the group undergoing CRT as compared to 25% (77/308) in the control group<xref ref-type="bibr" rid="B15">15</xref>. Therefore, the mortality rate in our study is within the range reported in large studies.</p>
      <p>In a CARE-HF substudy<xref ref-type="bibr" rid="B16">16</xref>, severe mitral regurgitation at 3 months was a predictor of total mortality, while in the study by Cabrera-Bueno et al., it was a predictor of worse clinical outcome and smaller reverse remodeling<xref ref-type="bibr" rid="B17">17</xref>. An Insync ICD substudy has not confirmed those findings for moderate mitral regurgitation<xref ref-type="bibr" rid="B18">18</xref>. Verheart et al<xref ref-type="bibr" rid="B19">19</xref>, in a study with 266 patients, have reported earlier significant ventricular remodeling in the group with moderate to severe mitral regurgitation. Regarding our study&#8217;s population, 46% had moderate to severe mitral regurgitation before implantation as compared to 28.5% in the first year (p &lt; 0.008 ), showing the beneficial effect of CRT in reducing mitral regurgitation and the high risk of severe mitral regurgitation 1 year after CRT.</p>
      <p>Diastolic dysfunction has shown to play a role in CRT, but with no value as an independent predictor of response in most studies. One physiopathological mechanism known to relate to the clinical improvement and better outcome of those patients concerns the reduction in diastolic dysfunction grade<xref ref-type="bibr" rid="B20">20</xref>. In that study, 41.5% of the patients had severe diastolic dysfunction (grades III and IV), while in the first and second years, only 13.5% and 21.4%, respectively. This can explain the clinical improvement of some patients with no correlation with remodeling or increase in EF<xref ref-type="bibr" rid="B21">21</xref>. In our study, 41.5% of the patients had grade III/IV diastolic dysfunction, and 13.5% in the follow-up (p &lt; 0.001). That differs from the findings of Salukhe et al<xref ref-type="bibr" rid="B22">22</xref>, who have reported neither clinical improvement nor remodeling in patients with severe diastolic dysfunction.</p>
      <p>An EF &#8804;30% 1 year after CRT implicated a 3.1-fold increase in the risk of cardiac death/Tx. The MIRACLE study<xref ref-type="bibr" rid="B23">23</xref> has reported a total 5.9% increase in EF in 6 months, and the CARE-HF study<xref ref-type="bibr" rid="B24">24</xref> has reported a 6.9% increase in 18 months, while, in our study, the median EF was 29% before implantation, 33% 1 year after, and 35% 2 years after implantation. Linde et al<xref ref-type="bibr" rid="B25">25</xref>, in a REVERSE study subanalysis<italic>,</italic> have shown that a baseline EF &lt; 30%, as compared to values of 30%-40%, related favorably to post-CRT survival, via an index comprising clinical and echocardiographic variables, while Kronborg et al<xref ref-type="bibr" rid="B26">26</xref> have shown that a baseline EF &lt; 22.5% increased post-CRT mortality. In a previous study, we observed that a baseline EF &lt; 25% correlated with worse cardiac outcome, which was not maintained after 1 year when assessed with other clinical variables<xref ref-type="bibr" rid="B27">27</xref>.</p>
      <p>In a study<xref ref-type="bibr" rid="B28">28</xref> with 65 chagasic patients with implantable automatic defibrillators, on a 40 &#177; 26.8-month follow-up, we observed annual mortality of 6.1% and no sudden death. In multivariate analysis, an EF &lt; 30% and low educational level were predictors of worse prognosis<xref ref-type="bibr" rid="B28">28</xref>. The present study had 11.2% of chagasic patients. In another study by our team, Chagas disease was related to an increase in the risk of death only on univariate analysis, probably exceeded by right ventricular dysfunction, a variable not assessed in this study<xref ref-type="bibr" rid="B27">27</xref>.</p>
      <p>The models were properly assessed by using proportionality tests and were internally validated, which increases the value of the data presented.</p>
      <p>Patients with at least 2 variables related with worse prognosis 1 year after CRT (grade III mitral regurgitation, grade III/IV diastolic dysfunction and EF &#8804;30%) should be assessed early in the search for other therapeutic options, such as electrode implantation in another left ventricular position, multisite left ventricular stimulation, mitral valve surgery, earlier cardiac Tx or artificial heart, considering the high cardiac mortality or need for emergency Tx of that population.</p>
      <sec>
        <title>Limitations</title>
        <p>This was a single-center study with a small sample and no intra- and inter-observer variability analysis between the echocardiographers. Several important echocardiographic variables were excluded.</p>
      </sec>
    </sec>
    <sec sec-type="conclusions">
      <title>Conclusion</title>
      <p>An EF &lt; 30%, severe diastolic dysfunction and severe mitral regurgitation 1 year after CRT indicate worse prognosis, and other therapeutic options should be considered in the presence of 2 of those variables.</p>
    </sec>
  </body>
  <back>
    <fn-group>
      <fn fn-type="supported-by">
        <p>
          <bold>Sources of Funding</bold>
        </p>
        <p>This study was funded by Capes and Funcap.</p>
      </fn>
      <fn fn-type="other">
        <p>
          <bold>Study Association</bold>
        </p>
        <p>This article is part of the thesis of Doctoral submitted by Eduardo Arrais Rocha, from Universidade de S&#227;o Paulo e Universidade Federal do Cear&#225;.</p>
      </fn>
    </fn-group>
    <ack>
      <title>Acknowledgements</title>
      <p>We thank professors Jos&#233; Wellington O. Lima, Luis Gustavo Bastos Pinho and Juv&#234;ncio Santos Nobre for statistical guidance, and Dr. &#205;talo Martins, coordinator of the post-graduation program of the inter-institutional doctorate (Dinter).</p>
    </ack>
    <ref-list>
      <title>References</title>
      <ref id="B01">
        <label>1</label>
        <mixed-citation>1 Bochi EA, Braga FG, Ferreira SM, Rohde LE, Oliveira WA, Almeida DR, et al; Sociedade Brasileira de Cardiologia. III Brazilian Guideline on chronic heart failure. Arq Bras Cardiol. 2009;93(1 Suppl 1):3-70.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Bochi</surname>
              <given-names>EA</given-names>
            </name>
            <name>
              <surname>Braga</surname>
              <given-names>FG</given-names>
            </name>
            <name>
              <surname>Ferreira</surname>
              <given-names>SM</given-names>
            </name>
            <name>
              <surname>Rohde</surname>
              <given-names>LE</given-names>
            </name>
            <name>
              <surname>Oliveira</surname>
              <given-names>WA</given-names>
            </name>
            <name>
              <surname>Almeida</surname>
              <given-names>DR</given-names>
            </name>
            <etal/>
            <collab>Sociedade Brasileira de Cardiologia</collab>
          </person-group>
          <article-title>III Brazilian Guideline on chronic heart failure</article-title>
          <source>Arq Bras Cardiol</source>
          <year>2009</year>
          <volume>93</volume>
          <issue>1</issue>
          <supplement>Suppl 1</supplement>
          <fpage>3</fpage>
          <lpage>70</lpage>
        </element-citation>
      </ref>
      <ref id="B02">
        <label>2</label>
        <mixed-citation>2 Tracy CM, Epstein AE, Darbar D, Dimarco JP, Dunbar SB, Estes NA 3rd, et al. 2012 ACCF/AHA/HRS focused update of the 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2012;60(14):1297-313.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Tracy</surname>
              <given-names>CM</given-names>
            </name>
            <name>
              <surname>Epstein</surname>
              <given-names>AE</given-names>
            </name>
            <name>
              <surname>Darbar</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Dimarco</surname>
              <given-names>JP</given-names>
            </name>
            <name>
              <surname>Dunbar</surname>
              <given-names>SB</given-names>
            </name>
            <name>
              <surname>Estes</surname>
              <given-names>NA 3rd</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>2012 ACCF/AHA/HRS focused update of the 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines</article-title>
          <source>J Am Coll Cardiol</source>
          <year>2012</year>
          <volume>60</volume>
          <issue>14</issue>
          <fpage>1297</fpage>
          <lpage>1313</lpage>
        </element-citation>
      </ref>
      <ref id="B03">
        <label>3</label>
        <mixed-citation>3 Saxon LA, Ellenbogen KA. Resynchronization therapy for the treatment of heart failure. Circulation. 2003;108(9):1044-8.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Saxon</surname>
              <given-names>LA</given-names>
            </name>
            <name>
              <surname>Ellenbogen</surname>
              <given-names>KA</given-names>
            </name>
          </person-group>
          <article-title>Resynchronization therapy for the treatment of heart failure</article-title>
          <source>Circulation</source>
          <year>2003</year>
          <volume>108</volume>
          <issue>9</issue>
          <fpage>1044</fpage>
          <lpage>1048</lpage>
        </element-citation>
      </ref>
      <ref id="B04">
        <label>4</label>
        <mixed-citation>4 Bax JJ, Bleeker GB, Marwick TH, Molhoek SG, Boersma E, Steendijk P, et al. Left ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy. J Am Coll Cardiol. 2004;44(9):1834-40.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Bax</surname>
              <given-names>JJ</given-names>
            </name>
            <name>
              <surname>Bleeker</surname>
              <given-names>GB</given-names>
            </name>
            <name>
              <surname>Marwick</surname>
              <given-names>TH</given-names>
            </name>
            <name>
              <surname>Molhoek</surname>
              <given-names>SG</given-names>
            </name>
            <name>
              <surname>Boersma</surname>
              <given-names>E</given-names>
            </name>
            <name>
              <surname>Steendijk</surname>
              <given-names>P</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Left ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy</article-title>
          <source>J Am Coll Cardiol</source>
          <year>2004</year>
          <volume>44</volume>
          <issue>9</issue>
          <fpage>1834</fpage>
          <lpage>1840</lpage>
        </element-citation>
      </ref>
      <ref id="B05">
        <label>5</label>
        <mixed-citation>5 Yu CM, Bleeker GB, Fung JW, Schalij MJ, Zhang Q, van der Wall EE, et al. Left ventricular reverse remodeling but not clinical improvement predicts long-term survival after cardiac resynchronization therapy. Circulation. 2005;112(11):1580-6.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Yu</surname>
              <given-names>CM</given-names>
            </name>
            <name>
              <surname>Bleeker</surname>
              <given-names>GB</given-names>
            </name>
            <name>
              <surname>Fung</surname>
              <given-names>JW</given-names>
            </name>
            <name>
              <surname>Schalij</surname>
              <given-names>MJ</given-names>
            </name>
            <name>
              <surname>Zhang</surname>
              <given-names>Q</given-names>
            </name>
            <name>
              <surname>van der Wall</surname>
              <given-names>EE</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Left ventricular reverse remodeling but not clinical improvement predicts long-term survival after cardiac resynchronization therapy</article-title>
          <source>Circulation</source>
          <year>2005</year>
          <volume>112</volume>
          <issue>11</issue>
          <fpage>1580</fpage>
          <lpage>1586</lpage>
        </element-citation>
      </ref>
      <ref id="B06">
        <label>6</label>
        <mixed-citation>6 Cleland JG, Ghio S. The determinants of clinical outcome and clinical response to CRT are not the same. Heart Fail Rev. 2012;17(6):755-66.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Cleland</surname>
              <given-names>JG</given-names>
            </name>
            <name>
              <surname>Ghio</surname>
              <given-names>S</given-names>
            </name>
          </person-group>
          <article-title>The determinants of clinical outcome and clinical response to CRT are not the same</article-title>
          <source>Heart Fail Rev</source>
          <year>2012</year>
          <volume>17</volume>
          <issue>6</issue>
          <fpage>755</fpage>
          <lpage>766</lpage>
        </element-citation>
      </ref>
      <ref id="B07">
        <label>7</label>
        <mixed-citation>7 Foley PW, Leyva F, Frenneaux MP. What is treatment success in cardiac resynchronization therapy? Europace. 2009;11(Suppl 5):v58-65.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Foley</surname>
              <given-names>PW</given-names>
            </name>
            <name>
              <surname>Leyva</surname>
              <given-names>F</given-names>
            </name>
            <name>
              <surname>Frenneaux</surname>
              <given-names>MP</given-names>
            </name>
          </person-group>
          <article-title>What is treatment success in cardiac resynchronization therapy?</article-title>
          <source>Europace</source>
          <year>2009</year>
          <volume>11</volume>
          <supplement>Suppl 5</supplement>
          <fpage>v58</fpage>
          <lpage>v65</lpage>
        </element-citation>
      </ref>
      <ref id="B08">
        <label>8</label>
        <mixed-citation>8 Camarozano A, Rabischoffsky A, Maciel BC, Brindeiro Filho D, Horowitz ES, Pena JL, et al; Sociedade Brasileira de Cardiologia. Diretrizes das indica&#231;&#245;es da ecocardiografia. Arq Bras Cardiol. 2009;93(6 supl. 3):e265-e302.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Camarozano</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Rabischoffsky</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Maciel</surname>
              <given-names>BC</given-names>
            </name>
            <name>
              <surname>Brindeiro</surname>
              <given-names>D</given-names>
              <suffix>Filho</suffix>
            </name>
            <name>
              <surname>Horowitz</surname>
              <given-names>ES</given-names>
            </name>
            <name>
              <surname>Pena</surname>
              <given-names>JL</given-names>
            </name>
            <etal/>
            <collab>Sociedade Brasileira de Cardiologia</collab>
          </person-group>
          <article-title>Diretrizes das indica&#231;&#245;es da ecocardiografia</article-title>
          <source>Arq Bras Cardiol</source>
          <year>2009</year>
          <volume>93</volume>
          <issue>6</issue>
          <supplement>supl. 3</supplement>
          <fpage>e265</fpage>
          <lpage>e302</lpage>
        </element-citation>
      </ref>
      <ref id="B09">
        <label>9</label>
        <mixed-citation>9 Gottdiener JS, Bednarz J, Devereux R, Gardin J, Klein A, Manning WJ, et al. American Society of Echocardiography recommendations for use of echocardiography in clinical trials. J Am Soc Echocardiogr. 2004;17(10):1086-119.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Gottdiener</surname>
              <given-names>JS</given-names>
            </name>
            <name>
              <surname>Bednarz</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Devereux</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Gardin</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Klein</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Manning</surname>
              <given-names>WJ</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>American Society of Echocardiography recommendations for use of echocardiography in clinical trials</article-title>
          <source>J Am Soc Echocardiogr</source>
          <year>2004</year>
          <volume>17</volume>
          <issue>10</issue>
          <fpage>1086</fpage>
          <lpage>1119</lpage>
        </element-citation>
      </ref>
      <ref id="B10">
        <label>10</label>
        <mixed-citation>10 Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. Eur J Echocardiogr. 2009;10(2):165-93.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Nagueh</surname>
              <given-names>SF</given-names>
            </name>
            <name>
              <surname>Appleton</surname>
              <given-names>CP</given-names>
            </name>
            <name>
              <surname>Gillebert</surname>
              <given-names>TC</given-names>
            </name>
            <name>
              <surname>Marino</surname>
              <given-names>PN</given-names>
            </name>
            <name>
              <surname>Oh</surname>
              <given-names>JK</given-names>
            </name>
            <name>
              <surname>Smiseth</surname>
              <given-names>OA</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Recommendations for the evaluation of left ventricular diastolic function by echocardiography</article-title>
          <source>Eur J Echocardiogr</source>
          <year>2009</year>
          <volume>10</volume>
          <issue>2</issue>
          <fpage>165</fpage>
          <lpage>193</lpage>
        </element-citation>
      </ref>
      <ref id="B11">
        <label>11</label>
        <mixed-citation>11 Boidol J, Sredniawa B, Kowalski O, Szulik M, Mazurek M, Sokal A, et al. Many response criteria are poor predictors of outcomes after cardiac resynchronization therapy: validation using data from the randomized trial. Europace. 2013;15(6):835-44.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Boidol</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Sredniawa</surname>
              <given-names>B</given-names>
            </name>
            <name>
              <surname>Kowalski</surname>
              <given-names>O</given-names>
            </name>
            <name>
              <surname>Szulik</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Mazurek</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Sokal</surname>
              <given-names>A</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Many response criteria are poor predictors of outcomes after cardiac resynchronization therapy: validation using data from the randomized trial</article-title>
          <source>Europace</source>
          <year>2013</year>
          <volume>15</volume>
          <issue>6</issue>
          <fpage>835</fpage>
          <lpage>844</lpage>
        </element-citation>
      </ref>
      <ref id="B12">
        <label>12</label>
        <mixed-citation>12 Vidal B, Delgado V, Mont L, Poyatos S, Silva E, Angeles Castel M, et al. Decreased likelihood of response to cardiac resynchronization in patients with severe heart failure. Eur J Heart Fail. 2010;12(3):283-7.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Vidal</surname>
              <given-names>B</given-names>
            </name>
            <name>
              <surname>Delgado</surname>
              <given-names>V</given-names>
            </name>
            <name>
              <surname>Mont</surname>
              <given-names>L</given-names>
            </name>
            <name>
              <surname>Poyatos</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Silva</surname>
              <given-names>E</given-names>
            </name>
            <name>
              <surname>Angeles Castel</surname>
              <given-names>M</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Decreased likelihood of response to cardiac resynchronization in patients with severe heart failure</article-title>
          <source>Eur J Heart Fail</source>
          <year>2010</year>
          <volume>12</volume>
          <issue>3</issue>
          <fpage>283</fpage>
          <lpage>287</lpage>
        </element-citation>
      </ref>
      <ref id="B13">
        <label>13</label>
        <mixed-citation>13 Aksoy H, Okutucu S, Kaya EB, Deveci OS, Evranos B, Aytemir K, et al. Clinical and echocardiographic correlates of improvement in left ventricular diastolic function after cardiac resynchronization therapy. Europace. 2010;12(9):1256-61.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Aksoy</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Okutucu</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Kaya</surname>
              <given-names>EB</given-names>
            </name>
            <name>
              <surname>Deveci</surname>
              <given-names>OS</given-names>
            </name>
            <name>
              <surname>Evranos</surname>
              <given-names>B</given-names>
            </name>
            <name>
              <surname>Aytemir</surname>
              <given-names>K</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Clinical and echocardiographic correlates of improvement in left ventricular diastolic function after cardiac resynchronization therapy</article-title>
          <source>Europace</source>
          <year>2010</year>
          <volume>12</volume>
          <issue>9</issue>
          <fpage>1256</fpage>
          <lpage>1261</lpage>
        </element-citation>
      </ref>
      <ref id="B14">
        <label>14</label>
        <mixed-citation>14 Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med. 2005;352(15):1539-49.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Cleland</surname>
              <given-names>JG</given-names>
            </name>
            <name>
              <surname>Daubert</surname>
              <given-names>JC</given-names>
            </name>
            <name>
              <surname>Erdmann</surname>
              <given-names>E</given-names>
            </name>
            <name>
              <surname>Freemantle</surname>
              <given-names>N</given-names>
            </name>
            <name>
              <surname>Gras</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Kappenberger</surname>
              <given-names>L</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>The effect of cardiac resynchronization on morbidity and mortality in heart failure</article-title>
          <source>N Engl J Med</source>
          <year>2005</year>
          <volume>352</volume>
          <issue>15</issue>
          <fpage>1539</fpage>
          <lpage>1549</lpage>
        </element-citation>
      </ref>
      <ref id="B15">
        <label>15</label>
        <mixed-citation>15 Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, et al. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med. 2004;350(21):2140-50.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Bristow</surname>
              <given-names>MR</given-names>
            </name>
            <name>
              <surname>Saxon</surname>
              <given-names>LA</given-names>
            </name>
            <name>
              <surname>Boehmer</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Krueger</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Kass</surname>
              <given-names>DA</given-names>
            </name>
            <name>
              <surname>De Marco</surname>
              <given-names>T</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure</article-title>
          <source>N Engl J Med</source>
          <year>2004</year>
          <volume>350</volume>
          <issue>21</issue>
          <fpage>2140</fpage>
          <lpage>2150</lpage>
        </element-citation>
      </ref>
      <ref id="B16">
        <label>16</label>
        <mixed-citation>16 Cleland J, Freemantle N, Ghio S, Fruhwald F, Shankar A, Marijanowski M, et al. Predicting the long-term effects of cardiac resynchronization therapy on mortality from baseline variables and the early response a report from the CARE-HF (Cardiac Resynchronization in Heart Failure) Trial. J Am Coll Cardiol. 2008;52(6):438-45.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Cleland</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Freemantle</surname>
              <given-names>N</given-names>
            </name>
            <name>
              <surname>Ghio</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Fruhwald</surname>
              <given-names>F</given-names>
            </name>
            <name>
              <surname>Shankar</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Marijanowski</surname>
              <given-names>M</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Predicting the long-term effects of cardiac resynchronization therapy on mortality from baseline variables and the early response a report from the CARE-HF (Cardiac Resynchronization in Heart Failure) Trial</article-title>
          <source>J Am Coll Cardiol</source>
          <year>2008</year>
          <volume>52</volume>
          <issue>6</issue>
          <fpage>438</fpage>
          <lpage>445</lpage>
        </element-citation>
      </ref>
      <ref id="B17">
        <label>17</label>
        <mixed-citation>17 Cabrera-Bueno F, Molina-Mora MJ, Alzueta J, Pena-Hernandez J, Jimenez-Navarro M, Fernandez-Pastor J, et al. Persistence of secondary mitral regurgitation and response to cardiac resynchronization therapy. Eur J Echocardiogr. 2010;11(2):131-7.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Cabrera-Bueno</surname>
              <given-names>F</given-names>
            </name>
            <name>
              <surname>Molina-Mora</surname>
              <given-names>MJ</given-names>
            </name>
            <name>
              <surname>Alzueta</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Pena-Hernandez</surname>
              <given-names>J</given-names>
            </name>
            <name>
              <surname>Jimenez-Navarro</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Fernandez-Pastor</surname>
              <given-names>J</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Persistence of secondary mitral regurgitation and response to cardiac resynchronization therapy</article-title>
          <source>Eur J Echocardiogr</source>
          <year>2010</year>
          <volume>11</volume>
          <issue>2</issue>
          <fpage>131</fpage>
          <lpage>137</lpage>
        </element-citation>
      </ref>
      <ref id="B18">
        <label>18</label>
        <mixed-citation>18 Boriani G, Gasparini M, Landolina M, Lunati M, Biffi M, Santini M, et al. Impact of mitral regurgitation on the outcome of patients treated with CRT-D: data from the InSync ICD Italian Registry. Pacing Clin Electrophysiol. 2012;35(2):146-54.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Boriani</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Gasparini</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Landolina</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Lunati</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Biffi</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Santini</surname>
              <given-names>M</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Impact of mitral regurgitation on the outcome of patients treated with CRT-D: data from the InSync ICD Italian Registry</article-title>
          <source>Pacing Clin Electrophysiol</source>
          <year>2012</year>
          <volume>35</volume>
          <issue>2</issue>
          <fpage>146</fpage>
          <lpage>154</lpage>
        </element-citation>
      </ref>
      <ref id="B19">
        <label>19</label>
        <mixed-citation>19 Verhaert D, Popovic Z, De S, Puntawangkoon C, Wolski K, Wilkoff B, et al. Impact of mitral regurgitation on reverse remodeling and outcome in patients undergoing cardiac resynchronization therapy. Circ Cardiovasc Imaging. 2012;5(1):21.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Verhaert</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Popovic</surname>
              <given-names>Z</given-names>
            </name>
            <name>
              <surname>De</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Puntawangkoon</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Wolski</surname>
              <given-names>K</given-names>
            </name>
            <name>
              <surname>Wilkoff</surname>
              <given-names>B</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Impact of mitral regurgitation on reverse remodeling and outcome in patients undergoing cardiac resynchronization therapy</article-title>
          <source>Circ Cardiovasc Imaging</source>
          <year>2012</year>
          <volume>5</volume>
          <issue>1</issue>
          <fpage>21</fpage>
        </element-citation>
      </ref>
      <ref id="B20">
        <label>20</label>
        <mixed-citation>20 Verbrugge FH, Verhaert D, Grieten L, Dupont M, Rivero-Ayerza M, De Vusser P, et al. Revisiting diastolic filling time as mechanistic insight for response to cardiac resynchronization therapy. Europace. 2013;15(12):1747-56.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Verbrugge</surname>
              <given-names>FH</given-names>
            </name>
            <name>
              <surname>Verhaert</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Grieten</surname>
              <given-names>L</given-names>
            </name>
            <name>
              <surname>Dupont</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Rivero-Ayerza</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>De Vusser</surname>
              <given-names>P</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Revisiting diastolic filling time as mechanistic insight for response to cardiac resynchronization therapy</article-title>
          <source>Europace</source>
          <year>2013</year>
          <volume>15</volume>
          <issue>12</issue>
          <fpage>1747</fpage>
          <lpage>1756</lpage>
        </element-citation>
      </ref>
      <ref id="B21">
        <label>21</label>
        <mixed-citation>21 Agacdiken A, Vural A, Ural D, Sahin T, Kozdag G, Kahraman G, et al. Effect of cardiac resynchronization therapy on left ventricular diastolic filling pattern in responder and nonresponder patients. Pacing Clin Electrophysiol. 2005;28(7):654-60.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Agacdiken</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Vural</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Ural</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Sahin</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Kozdag</surname>
              <given-names>G</given-names>
            </name>
            <name>
              <surname>Kahraman</surname>
              <given-names>G</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Effect of cardiac resynchronization therapy on left ventricular diastolic filling pattern in responder and nonresponder patients</article-title>
          <source>Pacing Clin Electrophysiol</source>
          <year>2005</year>
          <volume>28</volume>
          <issue>7</issue>
          <fpage>654</fpage>
          <lpage>660</lpage>
        </element-citation>
      </ref>
      <ref id="B22">
        <label>22</label>
        <mixed-citation>22 Salukhe TV, Francis DP, Clague JR, Sutton R, Poole-Wilson P, Henein MY. Chronic heart failure patients with restrictive LV filling pattern have significantly less benefit from cardiac resynchronization therapy than patients with late LV filling pattern. Int J Cardiol. 2005;100(1):5-12.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Salukhe</surname>
              <given-names>TV</given-names>
            </name>
            <name>
              <surname>Francis</surname>
              <given-names>DP</given-names>
            </name>
            <name>
              <surname>Clague</surname>
              <given-names>JR</given-names>
            </name>
            <name>
              <surname>Sutton</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Poole-Wilson</surname>
              <given-names>P</given-names>
            </name>
            <name>
              <surname>Henein</surname>
              <given-names>MY</given-names>
            </name>
          </person-group>
          <article-title>Chronic heart failure patients with restrictive LV filling pattern have significantly less benefit from cardiac resynchronization therapy than patients with late LV filling pattern</article-title>
          <source>Int J Cardiol</source>
          <year>2005</year>
          <volume>100</volume>
          <issue>1</issue>
          <fpage>5</fpage>
          <lpage>12</lpage>
        </element-citation>
      </ref>
      <ref id="B23">
        <label>23</label>
        <mixed-citation>23 St John Sutton MG, Plappert T, Abraham WT, Smith AL, DeLurgio DB, Leon AR, et al. Effect of cardiac resynchronization therapy on left ventricular size and function in chronic heart failure. Circulation. 2003;107(15):1985-90.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>St John Sutton</surname>
              <given-names>MG</given-names>
            </name>
            <name>
              <surname>Plappert</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Abraham</surname>
              <given-names>WT</given-names>
            </name>
            <name>
              <surname>Smith</surname>
              <given-names>AL</given-names>
            </name>
            <name>
              <surname>DeLurgio</surname>
              <given-names>DB</given-names>
            </name>
            <name>
              <surname>Leon</surname>
              <given-names>AR</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Effect of cardiac resynchronization therapy on left ventricular size and function in chronic heart failure</article-title>
          <source>Circulation</source>
          <year>2003</year>
          <volume>107</volume>
          <issue>15</issue>
          <fpage>1985</fpage>
          <lpage>1990</lpage>
        </element-citation>
      </ref>
      <ref id="B24">
        <label>24</label>
        <mixed-citation>24 Cleland JG, Daubert J-C, Erdmann E, Freemantle N, Gras D, Kappenberger L, et al. Longer-term effects of cardiac resynchronization therapy on mortality in heart failure [the CArdiac REsynchronization-Heart Failure (CARE-HF) trial extension phase]. Eur Heart J. 2006;27(16):1928-32.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Cleland</surname>
              <given-names>JG</given-names>
            </name>
            <name>
              <surname>Daubert</surname>
              <given-names>J-C</given-names>
            </name>
            <name>
              <surname>Erdmann</surname>
              <given-names>E</given-names>
            </name>
            <name>
              <surname>Freemantle</surname>
              <given-names>N</given-names>
            </name>
            <name>
              <surname>Gras</surname>
              <given-names>D</given-names>
            </name>
            <name>
              <surname>Kappenberger</surname>
              <given-names>L</given-names>
            </name>
            <etal/>
          </person-group>
          <article-title>Longer-term effects of cardiac resynchronization therapy on mortality in heart failure [the CArdiac REsynchronization-Heart Failure (CARE-HF) trial extension phase]</article-title>
          <source>Eur Heart J</source>
          <year>2006</year>
          <volume>27</volume>
          <issue>16</issue>
          <fpage>1928</fpage>
          <lpage>1932</lpage>
        </element-citation>
      </ref>
      <ref id="B25">
        <label>25</label>
        <mixed-citation>25 Linde C, Daubert C, Abraham WT, St John Sutton M, Ghio S, Hassager C, et al; REsynchronization reVErses Remodeling in Systolic left vEntricular dysfunction (REVERSE) Study Group. Impact of ejection fraction on the clinical response to cardiac resynchronization therapy in mild heart failure. Circ Heart Fail. 2013;6(6):1180-9.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Linde</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Daubert</surname>
              <given-names>C</given-names>
            </name>
            <name>
              <surname>Abraham</surname>
              <given-names>WT</given-names>
            </name>
            <name>
              <surname>St John Sutton</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Ghio</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Hassager</surname>
              <given-names>C</given-names>
            </name>
            <etal/>
            <collab>REsynchronization reVErses Remodeling in Systolic left vEntricular dysfunction (REVERSE) Study Group</collab>
          </person-group>
          <article-title>Impact of ejection fraction on the clinical response to cardiac resynchronization therapy in mild heart failure</article-title>
          <source>Circ Heart Fail</source>
          <year>2013</year>
          <volume>6</volume>
          <issue>6</issue>
          <fpage>1180</fpage>
          <lpage>1189</lpage>
        </element-citation>
      </ref>
      <ref id="B26">
        <label>26</label>
        <mixed-citation>26 Kronborg MB, Mortensen PT, Kirkfeldt RE, Nielsen JC. Very long term follow-up of cardiac resynchronization therapy: clinical outcome and predictors of mortality. Eur J Heart Fail. 2008;10(8):796-801.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Kronborg</surname>
              <given-names>MB</given-names>
            </name>
            <name>
              <surname>Mortensen</surname>
              <given-names>PT</given-names>
            </name>
            <name>
              <surname>Kirkfeldt</surname>
              <given-names>RE</given-names>
            </name>
            <name>
              <surname>Nielsen</surname>
              <given-names>JC</given-names>
            </name>
          </person-group>
          <article-title>Very long term follow-up of cardiac resynchronization therapy: clinical outcome and predictors of mortality</article-title>
          <source>Eur J Heart Fail</source>
          <year>2008</year>
          <volume>10</volume>
          <issue>8</issue>
          <fpage>796</fpage>
          <lpage>801</lpage>
        </element-citation>
      </ref>
      <ref id="B27">
        <label>27</label>
        <mixed-citation>27 Rocha EA, Pereira FT, Abreu JS, Lima JW, Monteiro MP, Rocha Neto AC, et al. Development and validation of predictive models of cardiac mortality and transplantation in resynchronization therapy. Arq Bras Cardiol. 2015; in press.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Rocha</surname>
              <given-names>EA</given-names>
            </name>
            <name>
              <surname>Pereira</surname>
              <given-names>FT</given-names>
            </name>
            <name>
              <surname>Abreu</surname>
              <given-names>JS</given-names>
            </name>
            <name>
              <surname>Lima</surname>
              <given-names>JW</given-names>
            </name>
            <name>
              <surname>Monteiro</surname>
              <given-names>MP</given-names>
            </name>
            <name>
              <surname>Rocha</surname>
              <given-names>AC</given-names>
              <suffix>Neto</suffix>
            </name>
            <etal/>
          </person-group>
          <article-title>Development and validation of predictive models of cardiac mortality and transplantation in resynchronization therapy</article-title>
          <source>Arq Bras Cardiol</source>
          <year>2015</year>
          <comment>in press</comment>
        </element-citation>
      </ref>
      <ref id="B28">
        <label>28</label>
        <mixed-citation>28 Pereira FT, Rocha EA, Monteiro MP, Rocha Neto AC, Daher EF, Rodrigues Sobrinho CR, et al. Long-term follow-up of patients with chronic Chagas disease and implantable cardioverter-defibrillator. Pacing Clin Electrophysiol. 2014;37(6):751-6.</mixed-citation>
        <element-citation publication-type="journal">
          <person-group person-group-type="author">
            <name>
              <surname>Pereira</surname>
              <given-names>FT</given-names>
            </name>
            <name>
              <surname>Rocha</surname>
              <given-names>EA</given-names>
            </name>
            <name>
              <surname>Monteiro</surname>
              <given-names>MP</given-names>
            </name>
            <name>
              <surname>Rocha</surname>
              <given-names>AC</given-names>
              <suffix>Neto</suffix>
            </name>
            <name>
              <surname>Daher</surname>
              <given-names>EF</given-names>
            </name>
            <name>
              <surname>Rodrigues</surname>
              <given-names>CR</given-names>
              <suffix>Sobrinho</suffix>
            </name>
            <etal/>
          </person-group>
          <article-title>Long-term follow-up of patients with chronic Chagas disease and implantable cardioverter-defibrillator</article-title>
          <source>Pacing Clin Electrophysiol</source>
          <year>2014</year>
          <volume>37</volume>
          <issue>6</issue>
          <fpage>751</fpage>
          <lpage>756</lpage>
        </element-citation>
      </ref>
    </ref-list>
  </back>
  <sub-article article-type="translation" id="SA1" xml:lang="pt">
    <front-stub>
      <title-group>
        <article-title>Par&#226;metros Ecocardiogr&#225;ficos Preditores de Pior Evolu&#231;&#227;o ap&#243;s a Terapia de Ressincroniza&#231;&#227;o Card&#237;aca</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <name>
            <surname>Rocha</surname>
            <given-names>Eduardo Arrais</given-names>
          </name>
          <xref ref-type="aff" rid="aff04">1</xref>
          <xref ref-type="aff" rid="aff05">2</xref>
          <xref ref-type="corresp" rid="c02"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Pereira</surname>
            <given-names>Francisca Tatiana Moreira</given-names>
          </name>
          <xref ref-type="aff" rid="aff05">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Abreu</surname>
            <given-names>Jos&#233; Sebasti&#227;o</given-names>
          </name>
          <xref ref-type="aff" rid="aff05">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Lima</surname>
            <given-names>Jos&#233; Wellington O.</given-names>
          </name>
          <xref ref-type="aff" rid="aff06">3</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Monteiro</surname>
            <given-names>Marcelo de Paula M.</given-names>
          </name>
          <xref ref-type="aff" rid="aff05">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Rocha</surname>
            <given-names>Almino Cavalcante</given-names>
            <suffix>Neto</suffix>
          </name>
          <xref ref-type="aff" rid="aff05">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Quidute</surname>
            <given-names>Ana Rosa Pinto</given-names>
          </name>
          <xref ref-type="aff" rid="aff05">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Go&#233;s</surname>
            <given-names>Camilla Viana A.</given-names>
          </name>
          <xref ref-type="aff" rid="aff04">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Rodrigues</surname>
            <given-names>Carlos Roberto Martins</given-names>
            <suffix>Sobrinho</suffix>
          </name>
          <xref ref-type="aff" rid="aff05">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Scanavacca</surname>
            <given-names>Maur&#237;cio Ibrahim</given-names>
          </name>
          <xref ref-type="aff" rid="aff04">1</xref>
        </contrib>
      </contrib-group>
      <aff id="aff04">
        <label>1</label>
        <institution content-type="orgname">Universidade de S&#227;o Paulo</institution>
        <addr-line>
          <named-content content-type="city">S&#227;o Paulo</named-content>
          <named-content content-type="state">SP</named-content>
        </addr-line>
        <country country="BR">Brazil</country>
        <institution content-type="original">Universidade de S&#227;o Paulo &#8211; USP, S&#227;o Paulo, SP</institution>
        <institution content-type="normalized">Universidade de S&#227;o Paulo</institution>
      </aff>
      <aff id="aff05">
        <label>2</label>
        <institution content-type="orgname">Universidade Federal do Cear&#225;</institution>
        <addr-line>
          <named-content content-type="city">Fortaleza</named-content>
          <named-content content-type="state">CE</named-content>
        </addr-line>
        <country country="BR">Brasil</country>
        <institution content-type="original">Universidade Federal do Cear&#225; &#8211; UFC, Fortaleza, CE &#8211; Brasil</institution>
        <institution content-type="normalized">Universidade Federal do Cear&#225;</institution>
      </aff>
      <aff id="aff06">
        <label>3</label>
        <institution content-type="orgname">Universidade Estadual do Cear&#225;</institution>
        <addr-line>
          <named-content content-type="city">Fortaleza</named-content>
          <named-content content-type="state">CE</named-content>
        </addr-line>
        <country country="BR">Brasil</country>
        <institution content-type="original">Universidade Estadual do Cear&#225; &#8211; UECE, Fortaleza, CE &#8211; Brasil</institution>
        <institution content-type="normalized">Universidade Estadual do Cear&#225;</institution>
      </aff>
      <author-notes>
        <corresp id="c02"><bold>Correspond&#234;ncia: Eduardo Arrais Rocha</bold> , Universidade Federal do Cear&#225;. Av. Padre Ant&#244;nio Tom&#225;s, 3535 / 1301, Coc&#243;. CEP 60192-120, Fortaleza, CE &#8211; Brasil. E-mail: <email>eduardoa@cardiol.br</email>, <email>eduardoarraisrocha@gmail.com</email>
					</corresp>
        <fn fn-type="con">
          <p>
            <bold>Contribui&#231;&#227;o dos autores</bold>
          </p>
          <p>Concep&#231;&#227;o e desenho da pesquisa: Rocha EA, Pereira FTM, Abreu JS, Lima JWO, Rocha Neto AC, Rodrigues Sobrinho CRM, Scanavacca MI; Obten&#231;&#227;o de dados: Rocha EA, Pereira FTM, Abreu JS, Monteiro MPM, Go&#233;s CVA; An&#225;lise e interpreta&#231;&#227;o dos dados: Rocha EA, Pereira FTM, Abreu JS, Lima JWO, Monteiro MPM, Rocha Neto AC; An&#225;lise estat&#237;stica: Rocha EA, Lima JWO, Quidute ARP; Obten&#231;&#227;o de financiamento: Rocha EA; Reda&#231;&#227;o do manuscrito: Rocha EA, Abreu JS, Rocha Neto AC, Quidute ARP, Rodrigues Sobrinho CRM, Scanavacca MI.</p>
        </fn>
        <fn fn-type="conflict">
          <p>
            <bold>Potencial conflito de interesse</bold>
          </p>
          <p>Declaro n&#227;o haver conflito de interesses pertinentes.</p>
        </fn>
      </author-notes>
      <abstract>
        <title>Resumo</title>
        <sec>
          <title>Fundamento:</title>
          <p>A terapia de ressincroniza&#231;&#227;o card&#237;aca (TRC) &#233; um tratamento recomendado pelas principais diretrizes mundiais. Entretanto, 30%-40% dos pacientes selecionados n&#227;o respondem ao tratamento.</p>
        </sec>
        <sec>
          <title>Objetivo:</title>
          <p>Elaborar um modelo ecocardiogr&#225;fico preditor de risco de &#243;bito card&#237;aco ou transplante (Tx) ap&#243;s 1 ano da TRC.</p>
        </sec>
        <sec>
          <title>M&#233;todo:</title>
          <p>Estudo observacional, prospectivo, com inclus&#227;o de 116 pacientes, sendo 69,8% do sexo masculino, com idade de 64,89 &#177; 11,18 anos, CF III (68,1%) e IV ambulatorial (31,9%), bloqueio de ramo esquerdo em 71,55%, e mediana da fra&#231;&#227;o de eje&#231;&#227;o (FE) de 29%. Avalia&#231;&#245;es foram feitas no per&#237;odo pr&#233;-implante e 6-12 meses ap&#243;s, e correlacionadas com mortalidade card&#237;aca/Tx no final do seguimento. Foram realizadas an&#225;lises de regress&#227;o de Cox e log&#237;stica, com a curva ROC e de sobrevida de Kaplan-Meier. O modelo foi validado internamente pelo &#8220;<italic>bootstrapping</italic>&#8221;.</p>
        </sec>
        <sec>
          <title>Resultados:</title>
          <p>Ocorreram 29 (25%) &#243;bitos/Tx durante o seguimento de 34,09 &#177; 17,9 meses. A mortalidade card&#237;aca/Tx foi de 16,3 %. No modelo multivariado de Cox, as vari&#225;veis FE &lt; 30%, disfun&#231;&#227;o diast&#243;lica grau III/IV e regurgita&#231;&#227;o mitral grau III, aferidas com 6-12 meses, relacionaram-se independentemente a aumento de mortalidade card&#237;aca ou Tx, com raz&#245;es de risco de 3,1, 4,63 e 7,11, respectivamente. A &#225;rea sob a curva ROC foi de 0,78.</p>
        </sec>
        <sec>
          <title>Conclus&#227;o:</title>
          <p>FE menor que 30%, disfun&#231;&#227;o diast&#243;lica grave e regurgita&#231;&#227;o mitral severa indicam pior progn&#243;stico ap&#243;s 1 ano da TRC, devendo ser consideradas outras op&#231;&#245;es terap&#234;uticas na presen&#231;a da combina&#231;&#227;o de duas dessas vari&#225;veis.</p>
        </sec>
      </abstract>
      <kwd-group xml:lang="pt">
        <title>Palavras-chave:</title>
        <kwd>Insufici&#234;ncia Card&#237;aca / mortalidade</kwd>
        <kwd>Ecocardiografia</kwd>
        <kwd>Marca-Passo Artificial</kwd>
        <kwd>Terapia de Ressincroniza&#231;&#227;o Card&#237;aca</kwd>
        <kwd>Fatores de Risco</kwd>
      </kwd-group>
    </front-stub>
    <body>
      <sec sec-type="intro">
        <title>Introdu&#231;&#227;o</title>
        <p>As evid&#234;ncias cient&#237;ficas colocam a terapia de ressincroniza&#231;&#227;o card&#237;aca (TRC) como classe I nas principais diretrizes de marca-passo e insufici&#234;ncia card&#237;aca congestiva (ICC) para os pacientes em tratamento cl&#237;nico otimizado, em ICC classe funcional (CF) II, III ou IV ambulatorial da <italic>New York Heart Association</italic> (NYHA), com fra&#231;&#227;o de eje&#231;&#227;o (FE) &#8804;35% e dist&#250;rbio de condu&#231;&#227;o intraventricular, principalmente no ramo esquerdo<xref ref-type="bibr" rid="B01">1</xref><sup>,</sup><xref ref-type="bibr" rid="B02">2</xref>.</p>
        <p>Entretanto, 30%-40% dos casos de TRC podem n&#227;o apresentar resultados favor&#225;veis, ou seja, pacientes submetidos a tratamento cir&#250;rgico, com riscos e custos altos, sem benef&#237;cios cl&#237;nicos, hemodin&#226;micos ou de sobrevida<xref ref-type="bibr" rid="B03">3</xref><sup>,</sup><xref ref-type="bibr" rid="B04">4</xref>.</p>
        <p>Diversos estudos procuraram identificar crit&#233;rios preditores de resposta &#224; TRC, mas com diferentes padr&#245;es de defini&#231;&#227;o de resposta, sendo utilizadas, predominantemente, compara&#231;&#245;es com o volume sist&#243;lico final (VSF) e melhora na CF<xref ref-type="bibr" rid="B05">5</xref>. Entretanto, os pacientes tidos como respondedores pelas an&#225;lises dos volumes ventriculares n&#227;o mant&#234;m correla&#231;&#227;o frequente com a melhora cl&#237;nica observada, teste de caminhada ou qualidade de vida. Essa discord&#226;ncia n&#227;o &#233; surpresa, considerando que a TRC atua por diversas vias hemodin&#226;micas e neuro-humorais<xref ref-type="bibr" rid="B06">6</xref>.</p>
        <p>Os efeitos da TRC envolvem modifica&#231;&#245;es no grau de regurgita&#231;&#227;o mitral, seja sist&#243;lica ou diast&#243;lica, no grau de disfun&#231;&#227;o diast&#243;lica, na fun&#231;&#227;o sist&#243;lica e no dissincronismo. Portanto, uma avalia&#231;&#227;o progn&#243;stica ou defini&#231;&#227;o de resposta que utilize apenas uma vari&#225;vel, como o VSF do ventr&#237;culo esquerdo (VE), &#233; pouco precisa.</p>
        <p>Assim, existe necessidade de cria&#231;&#227;o de &#237;ndices multifatoriais ou escores<xref ref-type="bibr" rid="B07">7</xref> para tentar, com maior acur&#225;cia, identificar os preditores de melhor sobrevida e para designar os verdadeiros respondedores. Esses &#237;ndices devem envolver um maior n&#250;mero de vari&#225;veis, com maiores taxas de sensibilidade e especificidade.</p>
        <p>O objetivo deste estudo foi elaborar um modelo ecocardiogr&#225;fico ap&#243;s a TRC, preditor de risco de &#243;bito card&#237;aco ou transplante (Tx) durante a evolu&#231;&#227;o.</p>
      </sec>
      <sec sec-type="methods">
        <title>M&#233;todos</title>
        <p>Trata-se de estudo observacional, prospectivo, analisando 116 pacientes com marca-passo multi-s&#237;tio implantado em hospital terci&#225;rio universit&#225;rio. As avalia&#231;&#245;es foram feitas pr&#233;-implante (1&#186; tempo de an&#225;lise) e com 1 ano (2&#186; tempo de an&#225;lise). As caracter&#237;sticas da popula&#231;&#227;o do estudo s&#227;o apresentadas na <xref ref-type="table" rid="t05">Tabela 1</xref>.</p>
        <table-wrap id="t05">
          <label>Tabela 1</label>
          <caption>
            <title>Descri&#231;&#227;o das caracter&#237;sticas basais da popula&#231;&#227;o do estudo</title>
          </caption>
          <table frame="hsides" rules="groups">
            <thead>
              <tr>
                <th align="left">
                  <bold>Vari&#225;veis Basais</bold>
                </th>
                <th align="center">
                  <bold>Resultados</bold>
                </th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td>Total de pacientes</td>
                <td align="center">116</td>
              </tr>
              <tr>
                <td>Sexo masculino</td>
                <td align="center">69,83%</td>
              </tr>
              <tr>
                <td>Idade</td>
                <td align="center">64,8 &#177; 11,1</td>
              </tr>
              <tr>
                <td>CF (NYHA) III</td>
                <td align="center">68,1%</td>
              </tr>
              <tr>
                <td>CF (NYHA) IV</td>
                <td align="center">31,9%</td>
              </tr>
              <tr>
                <td>Uso de betabloqueador</td>
                <td align="center">88,7%</td>
              </tr>
              <tr>
                <td>Uso de IECA</td>
                <td align="center">97,4%</td>
              </tr>
              <tr>
                <td>Uso de altas doses de diur&#233;ticos</td>
                <td align="center">31,9%</td>
              </tr>
              <tr>
                <td>N&#250;mero de interna&#231;&#245;es pr&#233;vias</td>
                <td align="center">108</td>
              </tr>
              <tr>
                <td>Cardiopatia chag&#225;sica</td>
                <td align="center">11,2%</td>
              </tr>
              <tr>
                <td>Cardiopatia isqu&#234;mica</td>
                <td align="center">29,3%</td>
              </tr>
              <tr>
                <td>Cardiopatia dilatada</td>
                <td align="center">59,4%</td>
              </tr>
              <tr>
                <td>Largura do QRS pr&#233;vio</td>
                <td align="center">160 ms</td>
              </tr>
              <tr>
                <td>BRE</td>
                <td align="center">71,50%</td>
              </tr>
              <tr>
                <td>FE (Simpson)</td>
                <td align="center">29%</td>
              </tr>
              <tr>
                <td>Disfun&#231;&#227;o diast&#243;lica grau III/ IV</td>
                <td align="center">41,5%</td>
              </tr>
              <tr>
                <td>RM grau II e III</td>
                <td align="center">46%</td>
              </tr>
              <tr>
                <td>DDVE</td>
                <td align="center">70 mm</td>
              </tr>
              <tr>
                <td>PA sist&#243;lica</td>
                <td align="center">115 &#177; 17 mmHg</td>
              </tr>
              <tr>
                <td>Veia p&#243;stero-lateral</td>
                <td align="center">45,4%</td>
              </tr>
              <tr>
                <td>Creatinina</td>
                <td align="center">1,1 mg/dL</td>
              </tr>
            </tbody>
          </table>
          <table-wrap-foot>
            <fn id="TFN05">
              <p>CF: Classe funcional da NYHA (New York Heart Association); IECA: Inibidor da enzima de convers&#227;o da angiotensina; BRE: Bloqueio de ramo esquerdo; FE: Fra&#231;&#227;o de eje&#231;&#227;o; RM: Regurgita&#231;&#227;o mitral; DDVE: Di&#226;metro diast&#243;lico do ventr&#237;culo esquerdo; PA: Press&#227;o arterial. As larguras do QRS pr&#233; e p&#243;s: a FE e o DDVE foram expressos em medianas (vari&#225;veis n&#227;o normais). A idade foi apresentada como m&#233;dia e desvio-padr&#227;o.</p>
            </fn>
          </table-wrap-foot>
        </table-wrap>
        <p>Os eletrodos do ventr&#237;culo direito (VD) foram posicionados preferencialmente em regi&#227;o apical (84%), tendo-se procurado manter maior dist&#226;ncia entre os eletrodos de VE e VD. Os modelos utilizados foram da St. Jude Medical em 92 pacientes, da Biotronik, em 12, da Medtronic, em 10, e da Guidant, em 2. Dois pacientes receberam implantes por mini-toracotomia. O n&#250;mero de pacientes selecionados foi baseado em estudos pr&#233;vios com desfechos semelhantes.</p>
        <p>As doses dos medicamentos foram modificadas durante o acompanhamento, conforme julgamento dos m&#233;dicos assistentes (3 profissionais especialistas em estimula&#231;&#227;o card&#237;aca da mesma equipe), os mesmos que realizaram todos os procedimentos cl&#237;nicos e cir&#250;rgicos. Seis vari&#225;veis ecocardiogr&#225;ficas consideradas de f&#225;cil aquisi&#231;&#227;o na pr&#225;tica di&#225;ria e de grande utilidade cl&#237;nica foram selecionadas para an&#225;lise.</p>
        <sec>
          <title>Par&#226;metros Ecocardiogr&#225;ficos</title>
          <p>Foram seguidas as normas das diretrizes brasileira<xref ref-type="bibr" rid="B08">8</xref> e americana de ecocardiografia para an&#225;lise dos diversos par&#226;metros ecocardiogr&#225;ficos<xref ref-type="bibr" rid="B09">9</xref>. Os exames ecocardiogr&#225;ficos foram realizados por tr&#234;s ecocardiografistas experientes, sendo 70% dos exames realizados pelo mesmo ecocardiografista. Foi utilizado o aparelho da GE, modelo Vivid 7, com transdutor 3.5-MHz. Os m&#233;dicos realizadores dos exames estavam blindados em rela&#231;&#227;o aos achados cl&#237;nicos e ecocardiogr&#225;ficos pr&#233;vios dos pacientes e tinham experi&#234;ncia na avalia&#231;&#227;o de pacientes, como os inclu&#237;dos no estudo.</p>
          <p>A an&#225;lise da fun&#231;&#227;o sist&#243;lica foi realizada pelo m&#233;todo de Simpson, no modo bidimensional duas e quatro c&#226;maras, seguido pela m&#233;dia. Os di&#226;metros ventriculares foram obtidos no modo M, segundo padroniza&#231;&#245;es das diretrizes<xref ref-type="bibr" rid="B09">9</xref>.</p>
          <p>A fun&#231;&#227;o diast&#243;lica foi avaliada pelo fluxo mitral (em repouso e ap&#243;s manobra de Valsalva), pelo Doppler tecidual e pela velocidade de propaga&#231;&#227;o do fluxo pelo modo-M em cores, sendo quatro os graus de disfun&#231;&#227;o diast&#243;lica: I = leve; II = moderado ou pseudonormal; III = acentuado ou padr&#227;o restritivo; IV = grave ou padr&#227;o restritivo irrevers&#237;vel<xref ref-type="bibr" rid="B10">10</xref>.</p>
          <p>O grau de regurgita&#231;&#227;o mitral foi avaliado atrav&#233;s do Doppler colorido, conforme o percentual de preenchimento do &#225;trio esquerdo. No refluxo leve, o percentual era menor que 20%, no moderado entre 20% e 40%, sendo valores acima desse percentual considerados importantes. Nesse contexto pr&#225;tico, o efeito Coanda foi interpretado como refluxo moderado quando restrito &#224; parede lateral do &#225;trio, e acentuado quando se estendia pelo polo superior do &#225;trio esquerdo.</p>
          <p>Realizou-se a aplica&#231;&#227;o do consentimento livre e esclarecido para todos os pacientes e o estudo foi aprovado pela comiss&#227;o de &#233;tica do hospital.</p>
        </sec>
        <sec>
          <title>An&#225;lise Estat&#237;stica</title>
          <p>Realizou-se o teste de normalidade de Shapiro-Wilk para classifica&#231;&#227;o das vari&#225;veis normais. As vari&#225;veis FE e di&#226;metro diast&#243;lico do ventr&#237;culo esquerdo (DDVE) n&#227;o tiveram distribui&#231;&#227;o normal.</p>
          <p>As vari&#225;veis categ&#243;ricas foram apresentadas como frequ&#234;ncias e porcentagens, e as vari&#225;veis cont&#237;nuas, como m&#233;dias e desvio-padr&#227;o ou medianas. As vari&#225;veis categ&#243;ricas foram comparadas pelo teste de Mac-Nemar, Stuart-Maxwell ou qui-quadrado. O teste <italic>t</italic> de Student foi utilizado para comparar distribui&#231;&#245;es de vari&#225;veis cont&#237;nuas, aproximadamente normais, e o teste de Wilcoxon/Mann-Whitney, para compara&#231;&#245;es de vari&#225;veis cont&#237;nuas, n&#227;o normais. Distribui&#231;&#245;es foram consideradas significativamente diferentes quando o valor p foi menor que 0,05.</p>
          <p>Avaliou-se a rela&#231;&#227;o univariada entre as vari&#225;veis ecocardiogr&#225;ficas e &#243;bito por doen&#231;as card&#237;acas ou Tx atrav&#233;s de curva de sobrevida de Kaplan-Meier, teste de <italic>log-rank</italic> e regress&#227;o de Cox. As vari&#225;veis cont&#237;nuas foram exploradas pela regress&#227;o de Cox na busca de um ponto de corte.</p>
          <p>Um modelo de regress&#227;o m&#250;ltipla de Cox foi desenvolvido no primeiro ano ap&#243;s a TRC para avaliar a contribui&#231;&#227;o independente de cada uma das vari&#225;veis ecocardiogr&#225;ficas previamente selecionadas com signific&#226;ncia no modelo univariado de Cox. As vari&#225;veis com valor de p &lt; 10% foram consideradas potenciais confundidoras. Cada uma das vari&#225;veis foi inclu&#237;da no modelo multivariado pela ordem decrescente do <italic>hazard</italic>, sendo exclu&#237;das quando o valor p foi maior ou igual a 5%.</p>
          <p>Foram realizadas an&#225;lises de regress&#227;o log&#237;stica, utilizando-se <italic>hazard</italic><xref ref-type="bibr" rid="B11">11</xref> como vari&#225;vel independente para medida de risco, e &#243;bito card&#237;aco/Tx como vari&#225;vel dependente. A acur&#225;cia dos modelos foi testada pela curva ROC, com suas sensibilidade e especificidade. O modelo foi elaborado pelas pontua&#231;&#245;es do <italic>hazard</italic>, dividido em categorias de risco de acordo com o n&#250;mero de vari&#225;veis presentes e classificadas em baixo, m&#233;dio e alto risco, respectivamente classes A, B e C. Para elabora&#231;&#227;o das pontua&#231;&#245;es, o <italic>hazard</italic> das vari&#225;veis individuais foi dividido pelo maior valor do <italic>hazard</italic> do modelo, multiplicado por 100 e arredondado para o algoritmo superior mais pr&#243;ximo.</p>
          <p>Para avaliar a proporcionalidade dos riscos associados com os fatores preditores, utilizou-se o teste de Schoenfeld. Para valida&#231;&#227;o interna do modelo, utilizou-se o m&#233;todo de <italic>bootstraping</italic>, sendo obtido um intervalo de confian&#231;a para os <italic>hazard</italic> estimados, gerando-se 10 mil amostras de tamanho 60, aleat&#243;rias, sem reposi&#231;&#227;o do conjunto de dados original. Para cada 10 mil amostras artificiais foram estimados os <italic>hazard</italic> correspondentes a cada covari&#225;vel. Ordenaram-se os valores para cada covari&#225;vel, sendo observado o intervalo de confian&#231;a de 95%.</p>
          <p>Os dados foram analisados pelo programa de computa&#231;&#227;o Stata/SE, vers&#227;o 12.1, (StataCorp LP, College Station, TX, EUA) e pelo programa &#8220;R&#8221; (<italic>R Foundation for Statistical Computing</italic>, Viena, &#193;ustria).</p>
        </sec>
      </sec>
      <sec sec-type="results">
        <title>Resultados</title>
        <p>Ocorreram 29 &#243;bitos durante o seguimento (25% de mortalidade total) em 34,09 &#177; 17,9 meses. A taxa de mortalidade card&#237;aca e Tx foi de 16,3% (19 pacientes), sendo que 6 desses pacientes foram transplantados durante o tempo do estudo, 5 devido a ICC refrat&#225;ria e 1 devido a tempestade arr&#237;tmica. Dos 6 pacientes transplantados, 3 faleceram devido ao est&#225;gio avan&#231;ado da doen&#231;a quando submetido ao Tx.</p>
        <p>No modelo multivariado de Cox, as vari&#225;veis FE &lt; 30%, disfun&#231;&#227;o diast&#243;lica e regurgita&#231;&#227;o mitral relacionaram-se independentemente a aumento de mortalidade card&#237;aca/Tx, com raz&#245;es de risco de 3,1; 4,63 e 7,11, respectivamente (<xref ref-type="table" rid="t06">Tabela 2</xref>).</p>
        <table-wrap id="t06">
          <label>Tabela 2</label>
          <caption>
            <title>Modelo com as vari&#225;veis ecocardiogr&#225;ficas no primeiro ano</title>
          </caption>
          <table frame="hsides" rules="groups">
            <thead>
              <tr>
                <th align="left">
                  <bold>Vari&#225;vel</bold>
                </th>
                <th align="center">
                  <bold>Hazard</bold>
                </th>
                <th align="center">
                  <bold>Valor de p</bold>
                </th>
                <th align="center">
                  <bold>IC 95%</bold>
                </th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td>RM</td>
                <td align="center">7,115132</td>
                <td align="center">0,001</td>
                <td align="center">2,26449 - 22,35604</td>
              </tr>
              <tr>
                <td>Disfun&#231;&#227;o diast&#243;lica</td>
                <td align="center">4,631782</td>
                <td align="center">0,004</td>
                <td align="center">1,631656 - 13,14824</td>
              </tr>
              <tr>
                <td>FE &lt; 30%</td>
                <td align="center">3,101647</td>
                <td align="center">0,035</td>
                <td align="center">1,083580 - 8,878182</td>
              </tr>
            </tbody>
          </table>
          <table-wrap-foot>
            <fn id="TFN06">
              <p>IC: Intervalo de confian&#231;a; Hazard: hazard ratio; RM: Regurgita&#231;&#227;o mitral grau III (severa) comparada aos graus II e I (moderada a leve); Disfun&#231;&#227;o diast&#243;lica - graus III e IV (disfun&#231;&#227;o severa) comparados com graus I e II (disfun&#231;&#227;o leve a moderada); FE: Fra&#231;&#227;o de eje&#231;&#227;o.</p>
            </fn>
          </table-wrap-foot>
        </table-wrap>
        <p>As vari&#225;veis significativas no modelo multivariado tamb&#233;m se mostraram significativas isoladamente no modelo de Kaplan-Meier, quando comparadas pelo teste de <italic>log-rank</italic> (p &lt; 0,001) (<xref ref-type="fig" rid="f05">Figuras 1</xref>-<xref ref-type="fig" rid="f07">3</xref>). A an&#225;lise do modelo pela curva <italic>ROC</italic> demonstrou uma &#225;rea sob a curva (AUC) de 0,785, com sensibilidade de 56,2%, especificidade de 94,1% e acur&#225;cia de 88,2%.</p>
        <fig id="f05">
          <label>Figura 1</label>
          <caption>
            <title>Curva de Kaplan-Meier para a vari&#225;vel fra&#231;&#227;o de eje&#231;&#227;o (FE), dicotomizada entre &gt; 30% e &#8804;30% no 2&#186; tempo (1&#186; ano) da an&#225;lise exclusivamente com as vari&#225;veis ecocardiogr&#225;ficas, com p &lt; 0,001, comparada pelo teste de log-rank.</title>
          </caption>
          <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf01-pt.jpg"/></fig>
        <fig id="f06">
          <label>Figura 2</label>
          <caption>
            <title>Curva de Kaplan-Meier para a vari&#225;vel disfun&#231;&#227;o diast&#243;lica (grau I e II comparado com grau III e IV) no 2&#186; tempo de an&#225;lise (1&#186; ano), com p &lt; 0,001, comparada pelo teste de log-rank.</title>
          </caption>
          <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf02-pt.jpg"/></fig>
        <fig id="f07">
          <label>Figura 3</label>
          <caption>
            <title>Curva de Kaplan-Meier para a vari&#225;vel regurgita&#231;&#227;o mitral (RM), sendo comparados os graus leve, moderado e severo no 2&#186; tempo de an&#225;lise (1&#186; ano), com p &lt; 0,001, pelo teste de log-rank.</title>
          </caption>
          <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf03-pt.jpg"/></fig>
        <p>Para os modelos propostos, foram testadas todas as vari&#225;veis quanto aos cumprimentos dos pressupostos de proporcionalidade dos riscos pelo teste de Schoenfeld (<xref ref-type="table" rid="t07">Tabela 3</xref>), com resultados que confirmaram o ajuste do modelo para as vari&#225;veis propostas. O intervalo de confian&#231;a de 95% foi obtido e confirmou a signific&#226;ncia estat&#237;stica das estimativas das propor&#231;&#245;es. O modelo foi, portanto, validado pelo m&#233;todo de <italic>bootstraping</italic> e n&#227;o mostrou aus&#234;ncia de ajuste ou exagerada sensibilidade dos dados.</p>
        <table-wrap id="t07">
          <label>Tabela 3</label>
          <caption>
            <title>Teste de proporcionalidade de risco com as vari&#225;veis ecocardiogr&#225;ficas</title>
          </caption>
          <table frame="hsides" rules="groups">
            <thead>
              <tr>
                <th align="left">
                  <bold>Vari&#225;vel</bold>
                </th>
                <th align="center">
                  <bold>H</bold>
                </th>
                <th align="center">
                  <bold>chi<sup>2</sup></bold>
                </th>
                <th align="center"/>
                <th align="center"/>
                <th align="center">
                  <bold>df</bold>
                </th>
                <th align="center"/>
                <th align="center"/>
                <th align="center">
                  <bold>Valor p</bold>
                </th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td>RM</td>
                <td align="center">-0,01674</td>
                <td align="center">0,00</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">1</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">0,9456</td>
              </tr>
              <tr>
                <td>Disfun&#231;&#227;o diast&#243;lica</td>
                <td align="center">0,43109</td>
                <td align="center">2,52</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">1</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">0,1127</td>
              </tr>
              <tr>
                <td>FE &lt; 30%</td>
                <td align="center">-0,09445</td>
                <td align="center">0,14</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">1</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">0,7096</td>
              </tr>
              <tr>
                <td>Resultado</td>
                <td align="center"/>
                <td align="center">2,84</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">3</td>
                <td align="center"/>
                <td align="center"/>
                <td align="center">0,4174</td>
              </tr>
            </tbody>
          </table>
          <table-wrap-foot>
            <fn id="TFN07">
              <p>Teste de Shoenfeld; H: Hazard basal; chi<sup>2</sup>: Qui-quadrado; df: Grau de liberdade; Valor p: N&#237;vel de signific&#226;ncia estat&#237;stica; RM: Regurgita&#231;&#227;o mitral; FE: Fra&#231;&#227;o de eje&#231;&#227;o.</p>
            </fn>
          </table-wrap-foot>
        </table-wrap>
        <p>A partir das combina&#231;&#245;es dessas vari&#225;veis, conseguimos elaborar um modelo e escore com 3 classes, sendo a classe A, de baixo risco de &#243;bito card&#237;aco/Tx, composta pela aus&#234;ncia das vari&#225;veis significativas na an&#225;lise multivariada, implicando em uma taxa livre de eventos (TLE) de 97,5% em 30 meses. A presen&#231;a de 1 vari&#225;vel (classe B) implicou em uma TLE de 83,1% em 30 meses, tendo a combina&#231;&#227;o de 2 ou 3 vari&#225;veis (classe C) implicado em TLE de 38,5% em 30 meses (<xref ref-type="fig" rid="f08">Figura 4</xref> e <xref ref-type="table" rid="t08">Tabela 4</xref>).</p>
        <fig id="f08">
          <label>Figura 4</label>
          <caption>
            <title>Modelo com as vari&#225;veis ecocardiogr&#225;ficas no 1&#186; ano: Classe A &#8211; Nenhuma vari&#225;vel (baixo risco de &#243;bito card&#237;aco ou transplante); Classe B &#8211; Presen&#231;a de uma vari&#225;vel (m&#233;dio risco); Classe C (alto risco) &#8211; presen&#231;a de 2 ou 3 vari&#225;veis (fra&#231;&#227;o de eje&#231;&#227;o &lt; 30%, disfun&#231;&#227;o diast&#243;lica grau III e IV contra grau I e II, e regurgita&#231;&#227;o mitral grau III contra grau II e I). A Classe A implica em uma taxa livre de eventos (TLE) de 97,5% em 30 meses, a Classe B implica em TLE de 83,1%, e a Classe C implica em TLE de 38,5% em 30 meses.</title>
          </caption>
          <graphic xlink:href="1678-4170-abc-105-06-0552-50108-gf04-pt.jpg"/></fig>
        <table-wrap id="t08">
          <label>Tabela 4</label>
          <caption>
            <title>Escore com as vari&#225;veis ecocardiogr&#225;ficas no primeiro ano</title>
          </caption>
          <table frame="hsides" rules="groups">
            <thead>
              <tr>
                <th align="left">
                  <bold>Vari&#225;vel</bold>
                </th>
                <th align="center">
                  <bold>
                    <italic>Hazard</italic>
                  </bold>
                </th>
                <th align="center">
                  <bold>N</bold>
                </th>
                <th align="center">
                  <bold>Pontos</bold>
                </th>
                <th align="center">
                  <bold>Classe</bold>
                </th>
                <th align="center">
                  <bold>Risco</bold>
                </th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td>Nenhuma</td>
                <td align="center">1,0</td>
                <td align="center">62</td>
                <td align="center">0</td>
                <td align="center">A</td>
                <td align="center">Baixo</td>
              </tr>
              <tr>
                <td>FE &lt; 30 %</td>
                <td align="center">3,1</td>
                <td align="center">20</td>
                <td align="center">3</td>
                <td align="center">B</td>
                <td align="center">Intermedi&#225;rio</td>
              </tr>
              <tr>
                <td>DD</td>
                <td align="center">4,6</td>
                <td align="center">3</td>
                <td align="center">5</td>
                <td align="center">B</td>
                <td align="center">Intermedi&#225;rio</td>
              </tr>
              <tr>
                <td>RM</td>
                <td align="center">7,1</td>
                <td align="center">3</td>
                <td align="center">7</td>
                <td align="center">B</td>
                <td align="center">Intermedi&#225;rio</td>
              </tr>
              <tr>
                <td>FE + DD</td>
                <td align="center">14,3</td>
                <td align="center">7</td>
                <td align="center">8</td>
                <td align="center">C</td>
                <td align="center">Alto</td>
              </tr>
              <tr>
                <td>FE + RM</td>
                <td align="center">22,0</td>
                <td align="center">4</td>
                <td align="center">10</td>
                <td align="center">C</td>
                <td align="center">Alto</td>
              </tr>
              <tr>
                <td>DD + RM</td>
                <td align="center">32,9</td>
                <td align="center">1</td>
                <td align="center">12</td>
                <td align="center">C</td>
                <td align="center">Alto</td>
              </tr>
              <tr>
                <td>FE + RM + DD</td>
                <td align="center">102,2</td>
                <td align="center">2</td>
                <td align="center">15</td>
                <td align="center">C</td>
                <td align="center">Alto</td>
              </tr>
            </tbody>
          </table>
          <table-wrap-foot>
            <fn id="TFN08">
              <p>HR: Hazard (propor&#231;&#227;o de risco); FE: Fra&#231;&#227;o de eje&#231;&#227;o; DD: Disfun&#231;&#227;o diast&#243;lica graus III e IV (severa) comparada a graus I e II (leve a moderada); RM: Regurgita&#231;&#227;o mitral grau III (severa) comparada aos graus II e I (moderada a leve); Classe A - categoria de baixo risco de &#243;bito card&#237;aco ou transplante; Classe B - risco intermedi&#225;rio de &#243;bito card&#237;aco ou transplante; Classe C - alto risco de &#243;bito card&#237;aco ou transplante. Utilizou-se hazard como vari&#225;vel independente no modelo de regress&#227;o log&#237;stica para elabora&#231;&#227;o do escore. A pontua&#231;&#227;o foi obtida pela propor&#231;&#227;o do hazard da vari&#225;vel, dividida pelo maior valor, multiplicado por 100 e arredondado para o n&#250;mero mais pr&#243;ximo. Os pontos das vari&#225;veis combinadas resultaram da soma de seus valores individuais.</p>
            </fn>
          </table-wrap-foot>
        </table-wrap>
      </sec>
      <sec sec-type="discussion">
        <title>Discuss&#227;o</title>
        <p>Tr&#234;s importantes vari&#225;veis ecocardiogr&#225;ficas (regurgita&#231;&#227;o mitral grau III<xref ref-type="bibr" rid="B12">12</xref>, disfun&#231;&#227;o diast&#243;lica grau III/IV<xref ref-type="bibr" rid="B13">13</xref> e FE &#8804;30%), quando presentes 1 ano ap&#243;s a TRC, mostraram-se preditoras de &#243;bito card&#237;aco ou Tx neste modelo ecocardiogr&#225;fico.</p>
        <p>O modelo foi obtido dentro de uma popula&#231;&#227;o considerada de alto risco, pois apresentava mediana da FE de 29%, DDVE de 70 mm, disfun&#231;&#227;o diast&#243;lica grave em 42% da amostra, regurgita&#231;&#227;o mitral moderada ou severa em 45,6%, e interna&#231;&#227;o por ICC no &#250;ltimo ano em 64%.</p>
        <p>A taxa de mortalidade geral do estudo foi de 25% (29/116) em 34 &#177; 17 meses, enquanto no trabalho de Yu e cols.<xref ref-type="bibr" rid="B05">5</xref> foi de 15,6% em 24 meses, sendo inclu&#237;dos nesse estudo pacientes com FE de 40% e em CF II, portanto, uma popula&#231;&#227;o de menor risco. No estudo <italic>CARE-HF</italic>, a mortalidade foi de 30% no grupo sem interven&#231;&#227;o, contra 20% no grupo submetido &#224; TRC em seguimento de 29,4 meses<xref ref-type="bibr" rid="B14">14</xref>. No estudo <italic>COMPANION</italic>, a taxa de &#243;bito foi de 21% (131/617) no grupo CRT, contra 25% (77/308) no grupo controle em seguimento de 24 meses<xref ref-type="bibr" rid="B15">15</xref>. Portanto, os dados de mortalidade do nosso estudo est&#227;o dentro da faixa descrita nos grandes trabalhos.</p>
        <p>Em um subestudo do <italic>CARE-HF</italic><xref ref-type="bibr" rid="B16">16</xref>, a regurgita&#231;&#227;o mitral severa com 3 meses foi preditora de mortalidade total, e no estudo de Cabrera-Bueno e cols., foi preditora de pior evolu&#231;&#227;o cl&#237;nica e de menor remodelamento reverso<xref ref-type="bibr" rid="B17">17</xref>. Um subestudo do <italic>Insync ICD</italic> n&#227;o confirmou tais achados para regurgita&#231;&#227;o mitral moderada<xref ref-type="bibr" rid="B18">18</xref>. Verheart e cols.<xref ref-type="bibr" rid="B19">19</xref>, em um estudo com 266 pacientes, relataram remodelamento ventricular importante mais precoce no grupo com regurgita&#231;&#227;o mitral moderada a severa. Na nossa popula&#231;&#227;o, 46% apresentavam regurgita&#231;&#227;o mitral moderada a severa pr&#233;-implante, contra 28,5% no primeiro ano (p &lt; 0,008), demonstrando efeito ben&#233;fico da TRC na redu&#231;&#227;o da regurgita&#231;&#227;o mitral e elevado risco da presen&#231;a de regurgita&#231;&#227;o mitral severa ap&#243;s 1 ano da TRC.</p>
        <p>O papel da disfun&#231;&#227;o diast&#243;lica na TRC tem sido demonstrado, entretanto sem valor como vari&#225;vel preditora independente de resposta na maioria dos estudos. Sabe-se que um dos mecanismos fisiopatol&#243;gicos relacionados a melhora cl&#237;nica e melhor evolu&#231;&#227;o desses pacientes est&#225; relacionado &#224; redu&#231;&#227;o no grau de disfun&#231;&#227;o diast&#243;lica<xref ref-type="bibr" rid="B20">20</xref>. Nesse estudo, 41,5% dos pacientes tinham disfun&#231;&#227;o diast&#243;lica grave (grau III e IV), enquanto no primeiro ano, apenas 13,5%, e no segundo ano, 21,4%. Isso pode explicar a melhora cl&#237;nica de alguns pacientes sem correla&#231;&#227;o com remodelamento ou com aumento na FE<xref ref-type="bibr" rid="B21">21</xref>. No nosso estudo, 41,5% dos pacientes apresentavam disfun&#231;&#227;o diast&#243;lica grau III/IV, contra 13,5% na evolu&#231;&#227;o (p &lt; 0,001), diferentemente do estudo de Salukhe e cols.<xref ref-type="bibr" rid="B22">22</xref>, que n&#227;o demonstraram melhora cl&#237;nica ou remodelamento nos pacientes com padr&#227;o de disfun&#231;&#227;o diast&#243;lica grave.</p>
        <p>O valor da FE &#8804;30% ap&#243;s 1 ano da TRC implicou em aumento de risco de 3,1 vezes de &#243;bito card&#237;aco/Tx. No estudo <italic>MIRACLE</italic><xref ref-type="bibr" rid="B23">23</xref>, houve aumento total em 6 meses de 5,9% na FE e, no estudo <italic>CARE-HF</italic><xref ref-type="bibr" rid="B24">24</xref>, de 6,9% em 18 meses, enquanto no nosso estudo a mediana da FE era de 29% pr&#233;-implante, 33% com 1 ano e 35 % com 2 anos ap&#243;s o implante. Linde e cols.<xref ref-type="bibr" rid="B25">25</xref>, em suban&#225;lise do estudo <italic>REVERSE,</italic> mostraram que a FE basal &lt; 30%, comparada com valores de 30%-40%, estava relacionada favoravelmente com a sobrevida ap&#243;s a TRC, atrav&#233;s de um &#237;ndice composto por vari&#225;veis cl&#237;nicas e ecocardiogr&#225;ficas, enquanto Kronborg e cols.<xref ref-type="bibr" rid="B26">26</xref>demonstraram que a FE basal &lt; 22,5% determinou aumento de mortalidade ap&#243;s TRC. Em estudo pr&#233;vio, observamos que a FE basal &lt; 25% teve correla&#231;&#227;o com pior evolu&#231;&#227;o card&#237;aca, entretanto n&#227;o mantida com 1 ano de evolu&#231;&#227;o quando avaliada em conjunto com outras vari&#225;veis cl&#237;nicas<xref ref-type="bibr" rid="B27">27</xref>.</p>
        <p>Em estudo<xref ref-type="bibr" rid="B28">28</xref> com 65 pacientes chag&#225;sicos com desfibriladores internos autom&#225;ticos, em seguimento de 40 &#177; 26,8 meses, observamos mortalidade anual de 6,1% e aus&#234;ncia de morte s&#250;bita. Na an&#225;lise multivariada, a FE &lt;30% e baixo n&#237;vel educacional foram preditores de pior progn&#243;stico<xref ref-type="bibr" rid="B28">28</xref>. No presente estudo, 11,2% eram chag&#225;sicos. Em outro estudo nosso, a doen&#231;a de Chagas esteve relacionada a aumento de risco de &#243;bito apenas na an&#225;lise univariada, provavelmente suplantada pela disfun&#231;&#227;o de VD, vari&#225;vel n&#227;o analisada neste estudo<xref ref-type="bibr" rid="B27">27</xref>.</p>
        <p>Os modelos foram adequadamente aferidos por testes de proporcionalidade e validados internamente, o que aumenta o valor dos dados apresentados.</p>
        <p>Pacientes com 2 ou mais vari&#225;veis (regurgita&#231;&#227;o mitral grau III, disfun&#231;&#227;o diast&#243;lica grau III/IV e FE &#8804;30%) relacionadas com pior progn&#243;stico ap&#243;s 1 ano da TRC devem ser avaliados precocemente em busca de outras op&#231;&#245;es terap&#234;uticas, como o implante de eletrodo em outra posi&#231;&#227;o de VE, estimula&#231;&#227;o em m&#250;ltiplos s&#237;tios de VE, cirurgia da v&#225;lvula mitral, Tx card&#237;aco mais precoce ou cora&#231;&#227;o artificial, considerando-se a elevada mortalidade card&#237;aca ou a necessidade de Tx de urg&#234;ncia desta popula&#231;&#227;o.</p>
        <sec>
          <title>Limita&#231;&#245;es</title>
          <p>O estudo foi unic&#234;ntrico, com uma amostra pequena e n&#227;o foram realizadas an&#225;lises de variabilidade intra- e inter-observador entre os m&#233;dicos ecocardiografistas. Foram exclu&#237;das diversas vari&#225;veis ecocardiogr&#225;ficas importantes.</p>
        </sec>
      </sec>
      <sec sec-type="conclusions">
        <title>Conclus&#227;o</title>
        <p>A FE menor que 30%, a disfun&#231;&#227;o diast&#243;lica grave e a regurgita&#231;&#227;o mitral severa ap&#243;s 1 ano da TRC indicam pior progn&#243;stico, devendo-se considerar outras op&#231;&#245;es terap&#234;uticas na presen&#231;a da combina&#231;&#227;o de 2 dessas vari&#225;veis.</p>
      </sec>
    </body>
    <back>
      <fn-group>
        <fn fn-type="supported-by">
          <p>
            <bold>Fontes de financiamento</bold>
          </p>
          <p>O presente estudo foi financiado pela Capes e Funcap.</p>
        </fn>
        <fn fn-type="other">
          <p>
            <bold>Vincula&#231;&#227;o acad&#234;mica</bold>
          </p>
          <p>Este artigo &#233; parte de tese de Doutorado de Eduardo Arrais Rocha pela Universidade de S&#227;o Paulo e Universidade Federal do Cear&#225;.</p>
        </fn>
      </fn-group>
      <ack>
        <title>Agradecimentos</title>
        <p>Somos gratos aos professores Jos&#233; Wellington O. Lima, Luis Gustavo Bastos Pinho e Juv&#234;ncio Santos Nobre pelas orienta&#231;&#245;es estat&#237;sticas, e tamb&#233;m ao Dr. &#205;talo Martins pela coordena&#231;&#227;o do programa de p&#243;s-gradua&#231;&#227;o do Dinter (doutorado interinstitucional).</p>
      </ack>
    </back>
  </sub-article>
</article>
