Prospective, randomized, double-blind study. To evaluate intravenous corticosteroids in preventing delayed extubation after multilevel corpectomy and strut graft reconstructive procedures and to identify risk factors for delayed extubation in these patients. We performed a prospective, randomized double-blind study in patients undergoing multilevel cervical corpectomy procedures. Our hypothesis was that high-dose perioperative steroids would decrease edema and thus decrease the incidence of delayed extubation. We studied patients undergoing 2- or 3-level anterior cervical corpectomy procedures with anterior strut graft reconstruction. Sixty-six patients were randomized to receive 3 doses of either intravenous dexamethasone (n = 35) or saline (n = 31). The first dose was given before the incision, with subsequent doses given 8 and 16 hours later. Patients remained intubated until postoperative day 1, at which time a cuff leak test was performed by the anesthesiology attending. If a leak was present, the patient was extubated. If not, the test was repeated each postoperative day until a leak was present, indicating a patent airway. Five of 35 (14%) in the steroid group and 6 of 31 (19%) in the saline group required delayed extubation (P = 0.22). There were no statistical differences in preoperative parameters of age, gender, diagnosis, smoking history, BMI, number of operative levels, or preoperative American Society of Anesthesiologists rating between the 2 groups. Similarly there were no differences between the groups for duration of anesthesia, intraoperative colloids or crystalloids, intraoperative blood loss, or intraoperative urine output. The data for both groups were pooled to evaluate risk factors for delayed extubation. The only statistically significant risk factor for delayed extubation in this study was female gender (P = 0.0001). Based on our data, we cannot recommend intravenous dexamethasone for prevention of delayed extubation after multilevel anterior cervical corpectomy and strut grafting procedures.
Read full abstract