• Martin Dres, Jean-Louis Teboul, Nadia Anguel, Laurent Guerin, Christian Richard, and Xavier Monnet.
• EA4533, Université Paris-Sud, Le Kremlin-Bicêtre, France, martin.dres@psl.aphp.fr.
• Intensive Care Med. 2015 Mar 1; 41 (3): 487-94.

PurposeWeaning-induced cardiac dysfunction is more likely to occur if the heart does not tolerate the changes in loading conditions induced by spontaneous breathing trial (SBT). We hypothesized that the presence of cardiac preload independence before an SBT is associated with weaning failure related to cardiac dysfunction.MethodsWe included 30 patients after a first failed 1-h T-tube SBT who had a transpulmonary thermodilution already in place. Preload independence [no increase in the pulse contour analysis-derived cardiac index ≥10 % during passive leg raising (PLR)] was assessed before the second SBT. Failure of the SBT related to cardiac dysfunction was defined by an increase in pulmonary artery occlusion pressure above 18 mmHg at the end of the SBT associated with clinical intolerance.ResultsFifty-seven SBTs were analyzed. The SBT failed in 46 cases. Overall, 31 failed SBTs were associated with weaning-induced cardiac dysfunction. During PLR, the cardiac index did not change in cases of failed SBTs with cardiac dysfunction, whereas it significantly increased in the other cases: 4 % (interquartile range, IQR 0-5) vs. 12 % (IQR 11-15), respectively. If PLR did not increase the cardiac index by more than 10 % before the SBT, the occurrence of SBT failure related to cardiac dysfunction was predicted with a sensitivity of 97 % [95 % confidence interval (CI) 83-100], specificity of 81 % (95 % CI 61-93) and area under the receiver-operating characteristic curve of 0.88 (95 % CI 0.78-0.98).ConclusionsPreload independence assessed by a negative PLR test performed before an SBT predicts weaning failure related to cardiac dysfunction.

28 keywords...

hide…