1996 OPEN FORUM Abstracts
AN AIRWAY CARE SCORE TO AID EXTUBATION AFTER ACUTE BRAIN INJURY
William M. Coplin, MD; Kathy D. Cooley, RRT; Gordon D. Rubenfeld, MD; David J. Pierson, MD, Harborview Medical Center, University of Washington, Seattle, WA
Background: Previous studies on the prediction of successful weaning have tended to lump together both discontinuation of mechanical ventilation and successful extubation as the end point. Acute brain injury is one clinical scenario where the needs for ventilatory support and endotracheal intubation may be separate: patients (pts) may fulfill weaning criteria yet still require airway protection against aspiration. This study sought to determine whether a six-part airway care score (ACS) could predict successful extubation once acute brain injury pts met standard spontaneous ventilatory parameters with adequate gas exchange.
Method: All intubated pts admitted to a Harborview intensive care unit over a 6 month period were screened for study entry. Significant extracranial trauma, organ failure, toxic or metabolic coma, and admission tracheotomy excluded pts. We identified the day when pts met standard weaning parameters and no longer had any indication for mechanical ventilation. There was daily recording (until hospital discharge, death, or tracheotomy) of the Glasgow Coma Scale (GCS), factors defining pneumonia (CDC criteria), and the ACS (sum of 6 components, from 0 [best] to 18 [worst]):
Spontaneous Cough Gag Sputum Quantity
0 Vigorous 0 Vigorous 0 None
1 Moderate 1 Moderate 1 1 pass
2 Weak 2 Weak 2 2 passes
3 None 3 None 3 >=3 passes
Sputum Viscosity Suctioning Frequency/8 hr Sputum Character
0 Watery 0 >3 hr 0 Clear
1 Frothy 1 q2-3 hr 1 Tan
2 Thick 2 q1-2 hr 2 Yellow
3 Tenacious 3 < q 1 hr 3 Green
Results: For 107 pts eventually extubated, median ACS when they first met theoretical extubation readiness criteria (SVP) was 8 (1-12); at the time of actual extubation (EXT) it was 8 (1-16). Coma (GCSo8) correlated with worse ACS scores (P=0.003). Eleven pts were reintubated for pulmonary/airway reasons. The ACS, at either SVP or EXT, did not predict successful extubation (P=NS). Two components predicted extubation success: spontaneous cough (P=0.014) and suction frequency (P=0.015), but no component or combination predicted reintubation in individual pts.
Conclusions: Spontaneous cough and suctioning frequency independently correlated with successful extubation in these brain-injured pts, but neither ACS nor its components could identify pts who would be reintubated. Whether ventilatory mechanics and other clinical data can improve prediction of extubation success once pts meet weaning criteria, and whether neurological status contributes to pneumonia and other complication, deserve further study.
Supported in part by NIH NS 30305 (WMC)