Abstract
OBJECTIVE: Airway obstruction after postoperative extubation is a dreaded but uncommon complication in patients undergoing circumferential cervical spine surgery (CCSS). The cuff leak test (CLT) has been utilized to assess air leak around the endotracheal tube which may reflect airway swelling. In this prospective observational study, we analyze the temporal evolution of CLT and perioperative factors that may influence it. METHODS: Twenty patients undergoing single-stage CCSS were managed according to our extubation protocol. Patients were maintained intubated overnight following surgery. They were extubated if a CLT > 200 mL and both intensive care unit (ICU) and Neurosurgery teams agreed that it was safe. Patients extubated in the first postoperative day (8 of 20) comprised the early group, and the remaining patients (12 of 20) the delayed group. Patient and operative data were analyzed as a single group and comparing both groups. RESULTS: The main indication for surgery was cervical deformity. Median number of levels fused was 5 anteriorly (range, 1-6) and 6 (range, 1-13) posteriorly. Patients were kept intubated for an average of 73.6 hours (range, 26-222 hours) and stayed in the ICU for 119.1 hours (range, 36-360 hours). There were 4 failed extubations and 3 patients (15%) required a tracheostomy. Patient profiles between both groups were very similar across most patient variables but differed significantly regarding infraglottic luminal area (p < 0.05). Patients with larger preoperative cuff leak values tended to have a shorter intubation period (p = 0.053). CONCLUSION: This study objectively demonstrates the difficulties in airway management following CCSS and provides useful insight for preoperative planning and counseling. Local anatomic factors influence airway outcome more than operative factors. The study format does not allow for testing of interventions but we suggest that patients with favorable anatomy (larger infraglottic luminal area) may benefit from a less strict airway management protocol.