Predicting the Need for Early Tracheostomy: A Multifactorial Analysis of 992 Intubated Trauma Patients

BACKGROUND: Tracheostomy has few, severe risks, while prolonged endotracheal intubation causes morbidity. The need for tracheostomy was assessed, based on early clinical parameters. METHODS: Adult trauma patients (January 1994-August 2004), intubated for resuscitation, ventilated >24 hours, were retrospectively evaluated for demographics, physiology, brain, and pulmonary injury. Tracheostomy patients were compared with those without. Chi-square, Mann-Whitney, and multivariate logistic regression were used with statistical significance at p < 0.05.* RESULTS: Of 992 patients, 430 (43%) underwent tracheostomy at 9.22 +/- 5.7 days. Risk factors were age (45.6* +/- 18.8 vs. 36.7 +/- 15.9, OR: 2.1 (18 years increments), ISS (30.3* +/- 12.5 vs. 22.0 +/- 10.3, OR: 2.1 (12u increments), damage control (DC) [68%*(n = 51) vs. 32%*(n = 51), OR: 3.8], craniotomy [70%*(n = 21) versus 30%(n = 9), OR: 2.6], and intracranial pressure monitor (ICP) [65.4%*(n = 87) vs. 34.6%(n = 46), OR: 2.1]. A 100% tracheostomy rate (n = 30, 3.0%) occurred with ISS (injury severity score) = 75, ISS >or=50, and age >or=55, admit/24 hour GCS (Glasgow Coma Scale) = 3 and age >or=70, AIS abdomen, chest or extremities >or=5 and age >or=60, bilateral pulmonary contusions (BPC) and >or=8 rib fractures, craniotomy and age >or=50, craniotomy with intracranial pressure (ICP) and age >or=40, or craniotomy and GCS or=90% (n = 105, 10.6%) was found with ISS >or=54, ISS >or=40, and age >or=40, admit/24 hour GCS = 3 and age >or=55, paralysis and age >or=40, BPC and age >or=55.A tracheostomy rate >or=80% (n = 248, 25.0%) occurred with ISS >or=38, age >or=80, admit/24 hour GCS = 3 and age >or=45, DC and age >or=50, BPC and age >or=50, aspiration and age >or=55, craniotomy with ICP, craniotomy with GCS or=90% risk undergo early tracheostomy and that it is considered in the >or=80% risk group to potentially decreased morbidity, increased patient comfort, and optimize resource utilization.