LAUSR

OBJECTIVES To determine the incidence of postoperative diaphragm dysfunction as diagnosed by ultrasonography. DESIGN Explorative prospective observational study. SETTING University intensive care unit. PARTICIPANTS One hundred consecutive patients undergoing elective cardiac surgery. INTERVENTIONS Diaphragm ultrasound was performed the day before surgery during unassisted breath (D-1), at the first spontaneous breathing trial attempt (DSBT), 24 hours after surgery (D+1), and at intensive care unit (ICU) discharge (DICU). Diaphragm displacement, inspiratory and expiratory thickness, and the thickening fraction were measured at all timepoints. MEASUREMENTS AND MAIN RESULTS Primary outcome was assessing the rate of postoperative diaphragm dysfunction, defined as a thickening fraction <20% at DSBT. Secondary outcomes were the number of difficult-to-wean patients, the need for rescue noninvasive ventilation, the reintubation rate, and the ICU length of stay. Thirty-eight patients showed diaphragm dysfunction at DSBT, which resolved over time. No differences in preoperative characteristics and comorbidities were found between patients who developed postoperative diaphragm dysfunction and patients without postoperative disorders. The duration of cardiopulmonary bypass (103 ± 34 v 55 ± 34 min; P < 0.001) was significantly associated with the development of postoperative diaphragm dysfunction. When compared with patients without postoperative diaphragm disorders, patients with diaphragm dysfunction were characterized by a higher rate of difficult weaning (32% v 5%; P < 0.001), lower extubation rate at 24 hours after surgery (50% v 92%; P < 0.001), and longer ICU length of stay (19 [16; 88] v 16 [15; 18] hours; P < 0.001). CONCLUSIONS The incidence of postoperative diaphragm dysfunction after elective cardiac surgery is high and might contribute to prolonging ICU length of stay.