A hard pill to swallow: factors associated with dysphagia following adult cervical deformity surgery

BACKGROUND CONTEXT: Dysphagia has long been intertwined with cervical interventions and significantly impacts patients’ quality of life. Still, the development and evolution of dysphagia is difficult to predict. PURPOSE: To discern which factors affect the development of postoperative dysphagia for preemptive identification of those most at risk. STUDY DESIGN/SETTING: Retrospective cohort study of prospectively enrolled database. PATIENT SAMPLE: A total of 705 adult cervical deformity patients. OUTCOME MEASURES: Complications, HRQLs. METHODS: Operative ACD patients with complete preoperative baseline (BL) and 2-year (2Y) data were analyzed via descriptive statistics. Patients who reported dysphagia as a postoperative complication or registered a SWAL-QOL score <25th percentile were categorized as the dysphagia cohort. Descriptive analyses, means comparison tests, cross tabulations, and regression analyses were run to register significant differences and associations between cohorts. RESULTS: A total of 265 patients met inclusion criteria (mean age 58.2±11.4 years 51.2% female, mean BL BMI 28.5±7.6 kg/m2, mean CCI: 0.93±1.3, mean BL frailty score). For the total cohort, in terms of surgical parameters, the mean operative time was 348±194.9 minutes, mean estimated blood loss (EBL) was 834.5±1180.3 mL, and mean levels fused was 5.9±3.6). Perioperatively, the mean length of stay (LOS) was 4.9±5.6 days. Eighty-two (30.9%) of patients reported postoperative dysphagia. At baseline, the dysphagia group demonstrated significantly greater BL frailty score (p<.001) and significantly greater BL C2-C7 (p=0.002) and cSVA(p=0.001). Those with morphologic focal kyphosis at BL were more likely to experience dysphagia (OR 4.3, p<.001). There was no significant difference amongst those who had previously had cervical surgery from any approach. Increasing operative time was also associated with a higher likelihood of acquiring dysphagia (OR 1.004, CI 95% 1.002-1.006, p<.001). Those who had an osteotomy from an anterior approach at any level from C3-C7 demonstrated significantly greater rates of dysphagia (p<.02, all). Degree of correction appeared to significantly influence the likelihood of dysphagia. Correction of both C2-C7 and MGS from severe → moderate/low were significant predictors of dysphagia (OR 3.6, p<.001; OR 17.0, p=0.009). Lastly, patients could not successfully be extubated immediately postoperatively demonstrated significantly greater occurrence of dysphagia (80.0% vs 20.0%, p=0.015); failure to extubate increased the odds of dysphagia 9.3x (OR 9.5, p=0.047). There was a significantly greater rate of dysphagia amongst those who experienced DJF by M3 (p=0.034). In terms of HRQLs, the dysphagia cohort reported a significantly greater M3 NSR Neck (p<.001). CONCLUSIONS: Dysphagia may occur in patients who exhibit greater frailty or deformity at baseline, those who undergo osteotomies from the mid-lower cervical region, and those who receive a significant degree of correction from baseline. Furthermore, delays in extubation can considerably increase the risk of dysphagia. FDA Device/Drug Status: This abstract does not discuss or include any applicable devices or drugs.

Duke Scholars

Author Christopher Ignatius Shaffrey Orthopaedic Surgery

Author Peter Passias Orthopaedic Surgery