Outcomes in single-level posterior cervical spine surgeries performed in the sitting and prone positions

OBJECTIVE: The sitting or semisitting position in neurosurgery allows for several technical advantages, including improved visualization of the surgical field. However, it has also been associated with an increased risk of venous air embolisms and positioning-related complications that limit its commonplace adoption. The authors report a large, single-center series of cervical spine procedures performed with patients in the sitting or prone position in order to assess the perceived risk of intraoperative and postoperative complications associated with the sitting position. METHODS: Noninstrumented, single-level posterior cervical spine procedures performed with patients in the sitting/ semisitting or prone position from 2000 to 2016 at a single institution were reviewed. Institutional abstraction tools (DataMart and Chart Plus) were used to collect data from the medical records. The two positions were compared with regard to preoperative factors, intraoperative variables, and postoperative outcomes. Multivariable logistic regression models were fitted for 30-day readmission, 30-day return to the operating room, and complication rates. RESULTS: A total of 750 patients (sitting, n = 480; prone, n = 270) were analyzed. The median age was 53 years for those who underwent surgery in the prone position and 50 years for those who underwent surgery in the sitting position (IQRs 45-62 years and 43-60 years, respectively), and 35% of the patients were female. Sitting cases were associated with significantly longer anesthetic times (221 minutes [range 199-252 minutes] vs 205 minutes [range 179-254 minutes]) and operative times (126 minutes [range 101-163 minutes] vs 149 minutes [120-181 minutes]). Cardiorespiratory events in the postanesthesia care unit (PACU) were comparable between the two groups, with the exception of episodes of apnea (2.6% vs 0.6%, p = 0.041) and hypoventilation (4.4% vs 0.8%, p < 0.003), which were more frequent in the prone-position cohort. On multivariable analysis, the effect of the sitting versus the prone position was not significant for 30-day readmission (OR 0.77, 95% CI 0.34-1.71, p = 0.52) or reoperation (OR 0.71, 95% CI 0.31-1.60, p = 0.40). The sitting position was associated with lower odds of developing any complication (OR 0.31, 95% CI 0.16-0.62, p < 0.001). CONCLUSIONS: Based on the intraoperative and postoperative complications chosen in this study, the sitting position confers a similar safety profile to the prone position. This can be explained by a more anatomic positioning accounting for reduced temporary neurological deficits and reduced PACU-associated hypoventilation noted in this series. Nevertheless, the findings may also reflect institutional familiarity, experience, and mastery of this position type, and outcomes may not reflect practices in general.