Study protocol
In the preoperative moment, all patients were functionally assessed using the FIM, six-minute walk test and peripheral muscle strength through the Medical Research Council (MRC).
The next day the patients were referred to the operating room. The surgery took place with the same surgical team, being performed via median sternotomy and cardiopulmonary bypass, using the graft of the external thoracic artery or bypass. All patients left the operating room with a subxiphoid and intercostal drain and were referred to the ICU with full analgesia. Upon arriving at this unit, they were conducted according to routine, with no influence from the researchers. In accordance with the weaning criteria, the patients were guided and, subsequently extubated, oxygen support at low flow was started, with a percentage sufficient to maintain saturation between 94 and 97%.
In the hospital where the research was carried out, NIV is routinely performed on the first postoperative day. Shortly after extubation, eligible patients were randomized using the electronic system: http://randomizer.org/form.htm, performed by a professional not belonging to the research group, and this procedure was kept confidential from other members of the search. Sealed, opaque and sequentially numbered envelopes were used to hide the allocation sequence until interventions were designated. The researchers responsible for the evaluations were blinded to the intervention and control groups.
Patients were divided into two groups: immediate NIV group (NIVI) and conventional NIV group (NIVC). The NIVI group performed noninvasive ventilation immediately after orotracheal extubation, whereas the NIVC group performed NIV on the first postoperative day around 24 hours after extubation.
NIV was performed on the Servo-S ventilator (Dräger Medical, Lübeck, Germany) in ventilation mode with support pressure, the pressure being sufficient to maintain the tidal volume between 6 to 8 ml/kg, positive pressure at the end of expiration starting at 5 cmH2O, reaching 12 cmH2O and inspired oxygen fraction of 30%. A face mask was used and the increase in PEEP was protocol, occurring in all patients. This therapy was maintained for 40 minutes in both groups and performed only once. An arterial blood gas analysis was collected before and after NIV for analysis of gas behavior. A day later, a new blood gas analysis was collected for late verification of oxygenation.
On the day of discharge from the ICU, patients were again evaluated for MIF, MRC and analyzed for pulmonary complications, deaths and length of stay in that unit. These assessments were repeated on the day of hospital discharge, adding up to the repetition of the six-minute walk test. All patients received standard assistance from the physiotherapy service, which consisted of kinesiotherapy, cycle ergometry and walking.
The outcomes related to postoperative complications were assessed by a radiologist who was blind. Gasometric and functional evaluation were performed one by a blind doctor and physiotherapist, respectively.