TY - JOUR AU - Wise, Eric Stephen AU - Stonko, David P. AU - Glaser, Zachary A. AU - Garcia, Kelly L. AU - Huang, Jennifer J. AU - Kim, Justine S. AU - Kallos, Justiss A. AU - Starnes, Joseph R. AU - Fleming, Jacob W. AU - Hocking, Kyle M. AU - Brophy, Colleen M. AU - Eagle, Susan S. PY - 2017/02/24 Y2 - 2024/03/29 TI - Prediction of Prolonged Ventilation after Coronary Artery Bypass Grafting: Data from an Artificial Neural Network JF - The Heart Surgery Forum JA - HSF VL - 20 IS - 1 SE - DO - 10.1532/hsf.1566 UR - https://journal.hsforum.com/index.php/HSF/article/view/1566 SP - E007-E014 AB - <p class="p1"><span class="s1"><strong>Objectives:</strong> The need for mechanical ventilation 24 hours after coronary artery bypass grafting (CABG) is considered a morbidity by the Society of Thoracic Surgeons. The purpose of this investigation was twofold: to identify simple preoperative patient factors independently associated with prolonged ventilation and to optimize prediction and early identification of patients prone to prolonged ventilation using an artificial neural network (ANN).</span></p><p class="p1"><span class="s1"><strong>Methods:</strong> Using the institutional Adult Cardiac Database, 738 patients who underwent CABG since 2005 were reviewed for preoperative factors independently associated with prolonged postoperative ventilation. Prediction of prolonged ventilation from the identified variables was modeled using both “traditional” multiple logistic regression and an ANN. The two models were compared using Pearson <em>r</em><sup>2</sup> and area under the curve (AUC) parameters.</span></p><p class="p1"><span class="s1"><strong>Results:</strong> Of 738 included patients, 14% (104/738) required mechanical ventilation </span><span class="s2">≥</span><span class="s1"> 24 hours postoperatively. Upon multivariate analysis, higher body-mass index (BMI; odds ratio [OR] 1.10 per unit, <em>P</em> &lt; 0.001), lower ejection fraction (OR 0.97 per %, <em>P</em> = 0.01) and use of cardiopulmonary bypass (OR 2.59, <em>P</em> = 0.02) were independently predictive of prolonged ventilation. The Pearson <em>r</em><sup>2</sup> and AUC of the multivariate nominal logistic regression model were 0.086 and 0.698 ± 0.05, respectively; analogous statistics of the ANN model were 0.159 and 0.732 ± 0.05, respectively.</span></p><p class="p1"><span class="s1">BMI, ejection fraction and cardiopulmonary bypass represent three simple factors that may predict prolonged ventilation after CABG. Early identification of these patients can be optimized using an ANN, an emerging paradigm for clinical outcomes modeling that may consider complex relationships among these variables.</span></p> ER -