Predictive abilities of O-C2a and O-EAa for the development of postoperative dysphagia in patients undergoing occipitocervical fusion

Abstract

Background ContextDysphagia is a common postoperative complication in patients undergoing occipitocervical fusion (OCF). Previous studies had proposed the use of two measures—the occipital to C2 angle (O-C2a) and the occipital and external acoustic meatus to axis angle (O-EAa)—to predict postoperative dysphagia after OCF. However, these studies had small sample sizes and the predictive abilities of both measures are still not clear. PurposeTo evaluate the predictive ability of O-EAa and O-C2a for dysphagia after OCF. Study DesignA retrospective clinical study. Patient SampleA total of 109 consecutive patients who had undergone OCF. Outcome MeasuresPresence of postoperative dysphagia, O-C2a, C2 tilting angle (C2Ta), O-EAa, and the narrowest oropharyngeal airway space (nPAS). MethodsBetween April 2010 and June 2018, 109 consecutive patients who had undergone OCF were reviewed. Patients were divided into two groups according to the presence of postoperative dysphagia. Radiographic measurements, including O-C2a, C2Ta, O-EAa, and nPAS, were evaluated at preoperative and 1 month postoperative and the findings were compared. Simple linear regression was used to measure the correlations between the parameters and the presence of dysphagia, and the correlations within the parameters. Multiple regression analysis was used to examine the variables that affected the change of nPAS (dnPAS%). Sensitivity and specificity analyses were used to evaluate the effectiveness of the previously proposed measures (“O-C2a change≤−5°” and “postoperative O-EAa<100°”) for prediction of post-OCF dysphagia. ResultsThe incidence of dysphagia after OCF was 26.6% (29/109). Preoperative values for the radiographic parameters were similar between patients with and without dysphagia. In the dysphagia group, both O-C2a and O-EAa values showed a dramatic decrease after surgery, which was accompanied by a decrease in nPAS. Postoperative O-C2a, O-EAa, and nPAS in the dysphagia group were significantly smaller than those in the nondysphagia group (p<.05). The changes in O-EAa, O-C2a, and nPAS showed a linear correlation with the presence of dysphagia (p<.05). In addition, linear correlations were found between two of the three parameters. Multiple regression showed the change of O-C2a and O-EAa were significant predictors for dnPAS% (β=0.200, p=.022 and β=0.549, p=.000). The sensitivity and specificity of “O-C2a change≤−5°” in predicting dysphagia were 75.9% and 80.0% respectively, and those of “postoperative O-EAa<100°” were 75.9% and 62.5%, respectively. However, the sensitivity of the combination of these two values in predicting postoperative dysphagia was as high as 96.6%. ConclusionBoth O-EAa and O-C2a could be critical predictors for postoperative dysphagia. During surgery, ensuring that the O-EAa exceeds 100° and simultaneously avoiding an O-C2a reduction greater than 5° could effectively avert postoperative dysphagia.