Prone Positioning in Cardiac Surgery: For Many, But Not for Everyone

• Published in: Seminars in thoracic and cardiovascular surgery 2016, Vol.28 (2), p.281-287
• Main Authors: von Wardenburg, Che, MD, Wenzl, Martin, MD, Dell’Aquila, Angelo M., MD, Junger, Axel, MD, Fischlein, Theodor, MD, Santarpino, Giuseppe, MD
• Format: Article
• Language: English
• Subjects: Acute respiratory distress syndrome; Aged; Aged, 80 and over; ARDS; Cardiac Surgical Procedures - adverse effects; Cardiac Surgical Procedures - mortality; Cardiothoracic Surgery; Female; Germany; Hospital Mortality; Humans; Hypoxia - diagnosis; Hypoxia - etiology; Hypoxia - mortality; Hypoxia - therapy; Logistic Models; Male; Middle Aged; Multivariate Analysis; Odds Ratio; Outcomes; Patient Positioning - adverse effects; Patient Positioning - methods; Patient Positioning - mortality; Patient Selection; Postoperative care; Prone Position; Respiratory Distress Syndrome, Adult - diagnosis; Respiratory Distress Syndrome, Adult - etiology; Respiratory Distress Syndrome, Adult - mortality; Respiratory Distress Syndrome, Adult - therapy; Respiratory Insufficiency - diagnosis; Respiratory Insufficiency - etiology; Respiratory Insufficiency - mortality; Respiratory Insufficiency - therapy; Respiratory therapy; Retrospective Studies; Risk Factors; Time Factors; Treatment Outcome; Ventilation
• ISSN: 1043-0679; 1532-9488
• DOI: 10.1053/j.semtcvs.2016.04.008

Objective Prone positioning is a therapeutic maneuver to improve arterial oxygenation in patients with acute lung injury that is not implemented in most centers performing adult cardiac surgery. The aim of this study was to review our experience with prone positioning to assess the effects of this maneuver in patients with postoperative acute respiratory failure. Methods From 2010 to 2014, 127 adult patients with postoperative acute respiratory failure were treated with prone positioning in addition to specific therapy. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors associated with in-hospital mortality. Results In-hospital mortality was 22.8% ( n =29). No significant differences were observed in preoperative risk factors between patients who survived (S) and those who died (D), except for age (62.7±11.2 vs 70.2±11.3; p =0.007 - at multivariate analysis p =0.03, OR 1.1/year). Pre-proning values of PaO2 /FiO2 were significantly different between groups (D vs S: 115±46 vs 150±56; p =0.006), but only pre-proning FiO2 remained highly significant at multivariate analysis (D vs S: 0.82±0.18 vs 0.67±0.16; p =0.001, OR1.07; with FiO2 >0.75 vs