Objective: Prolonged mechanical ventilation after cardiac surgery is a serious complication that warrants search for new treatment strategies. Our objective was to identify patients still requiring mechanical ventilation 3 days after the operation and those successfully weaned by day 10 to avoid needless and potentially hazardous interventions, such as tracheostomy.
Methods: All consecutive patients still mechanically ventilated on day 3 after cardiac surgery were included in a prospective observational cohort. Patients' preoperative, intraoperative, and postoperative data were recorded. Logistic regression analysis was used to identify factors associated with successful weaning from mechanical ventilation by postoperative day 10.
Results: Among 2620 patients who underwent cardiac surgery, 163 were still receiving ventilatory assistance on day 3. By day 10, 50 (31%) patients had been successfully weaned, 78 (48%) were still receiving mechanical ventilation, and 35 (21%) had died. Multivariable regression analysis retained 6 day-3 factors associated with successful weaning (odds ratio): urine output 500 mL/24 hours or greater (16.47), Glasgow coma score of 15 (9.75), arterial bicarbonates 20 mmol/L or greater (6.09), platelet count 100 g/L or greater (3.18), patients without inotropic support with epinephrine/norepinephrine (2.84), and absence of lung injury (2.40). The area under the receiver operating characteristics curve for the simple score based on this model's beta-coefficients was 0.84 (95% confidence intervals, 0.78-0.91). Depending on the threshold chosen for this scoring system, only 3% to 17% of the patients would have received a needless intervention.
Conclusions: A simple score based on postoperative day-3 physiologic parameters might help intensivists early identify patients with a strong likelihood of success in rapid weaning from mechanical ventilation and therefore prevent needless procedures aimed at reducing duration of mechanical ventilation and related complications.