Abstract
Introduction
Dysphagia is one of the most common postoperative complaints following anterior cervical spine surgery. Dysphagia has been associated with increased postoperative morbidity including increased hospital length of stay, higher rates of readmission, and functional disability. 1 While this issue is undoubtedly multifactorial in nature, some of the proposed etiologies include prolonged esophageal retraction resulting in neuropraxia to the esophageal branches of the recurrent laryngeal nerve, esophageal ischemia, and local edema, particularly in the setting of prior neck surgery. 2 With that being said, patient-specific risk factors have yet to be fully elucidated in the literature as numerous studies offer conflicting conclusions.1-3
Although the issue of dysphagia is a well-documented perioperative issue following anterior cervical spine surgery, and more specifically anterior cervical discectomy and fusion (ACDF), several studies evaluating the incidence and severity of this issue have been limited by their retrospective designs, cohort sizes, and inconsistent use of dysphagia assessment tools. Therefore, the reported severity of dysphagia and associated degree of impairment that occurs following ACDF also varies significantly within the literature, with most studies suggesting that symptoms are transient and generally resolve without treatment.3-6 In addition, the impact of anterior cervical surgery on patients with baseline dysphagia also remains unclear. While the literature does highlight that preoperative dysphagia is a risk-factor for developing or retaining postoperative dysphagia, it is unclear if this finding has any significant impact on a patient’s immediate postoperative course. Therefore, the purpose of this study was to prospectively evaluate the impact of preoperative dysphagia on the postoperative incidence and severity of dysphagia in patients undergoing ACDF at multiple institutions. We hypothesize that preoperative dysphagia increases the postoperative incidence and severity of dysphagia in patients following ACDF.
Materials and Methods
Patient Selection and Data Collection
After Institutional Review Board (IRB) approval was obtained, all subjects provided informed written consent to participate in the study. Patients over 18 years of age who underwent an elective anterior discectomy and fusion (ACDF) for degenerative conditions were prospectively enrolled at two academic centers. Patients were excluded from the study if they received surgical intervention for tumors, infections, trauma, or revision at the index level. Informed consent for study participation was obtained, and patients were asked to complete a preoperative demographic and medical history questionnaire that included age, sex, BMI, smoking status (never, former, current smoker), alcohol consumption, presence of gastroesophageal reflux disease (GERD) and major depressive disorder (MDD), and preoperative usage of a proton-pump inhibitor (PPI) and/or selective serotonin reuptake inhibitor (SSRI). In addition, postoperative information was obtained via chart review that included number of levels fused, surgery length, perioperative IV corticosteroid use, and clinical follow-up. Preoperative dysphagia was self-reported by patients through a pre-operative questionnaire on a binary basis (yes or no). The cause of pre-operative dysphagia was not evaluated. Patients also completed dysphagia surveys (Bazaz, Dysphagia Short Questionnaire, 10-item Eating Assessment Tool) to assess dysphagia severity during their preoperative visit, and these dysphagia surveys were repeated immediately postoperatively, at two weeks and again at six, 12, and 24 weeks postoperatively during regularly scheduled follow-up appointments. Additionally, in-hospital course events were recorded including diet alteration, unplanned postoperative steroid usage, consults from speech pathology and otolaryngology (ENT), adjunct cervical spine magnetic resonance imaging (MRI), nasogastric tube usage, and reintubation.
Dysphagia Questionnaires
Statistical Analysis
Patients were grouped according to preoperative dysphagia status. All data points were entered into and stored in a computerized database. Descriptive statistics were used to present patient demographics and outcome measures, and were characterized by mean ± standard deviation. Continuous variables were analyzed using either an independent t-test or Mann Whitney U-test, for parametric and non-parametric data, respectively. All categorical variables were compared using a Chi-square or Fisher’s Exact test depending on cell count. A subset of the patient cohort with additional data for evaluation was subjectively analyzed for further consideration of in-hospital events or outcomes and compared in the same manner. Additionally, patients were stratified into three subgroups based on dysphagia status preoperatively and immediately postoperatively (patients with preoperative dysphagia, patients without preoperative dysphagia who developed postoperative dysphagia, patients without preoperative dysphagia who did not develop postoperative dysphagia) and compared using ANOVA tests or Kruskal–Wallis tests for continuous variables and Pearson chi-square analysis or Fisher’s Exact test for categorical variables. Statistical analysis utilized R Studio Version 4.0.2 (Boston, MA). A
Results
Patient Demographics and Surgical Characteristics
Demographics of Cohort.
Pre- and Post-Operative Dysphagia Characteristics and Outcomes
Dysphagia Characteristics.
Post-Operative In-Hospital Events and Outcomes
In Hospital and Postoperative Events.
Sub-Analysis of Post-Operative Dysphagia Outcomes and In-Hospital Events
Sub-analysis of Postoperative Dysphagia Outcomes and In-Hospital Events.
When evaluating the incidence of postoperative dysphagia over the 24-week post-operative period, patients with new-onset post-operative dysphagia reported a higher rate of dysphagia incidence at six weeks (58.3%,
Discussion
Postoperative dysphagia is common after ACDF and can be an unavoidable outcome associated with the procedure. Symptoms are generally only present in the early postoperative period but may continue into the 12-week postoperative period. 7 Reported incidence varies in the literature, ranging from 1.7 to 71% postoperatively, 8 with the wide variation at least in part due to differences in how dysphagia is defined within individual investigations. Despite such high dysphagia rates reported, risk factors for developing dysphagia after ACDF are incompletely understood. Potential causes are multifactorial and may be attributed to the thickness of the cervical plate, soft tissue swelling, and reperfusion injury. Additionally, the number of levels involved in the ACDF may contribute due to the amount of dissection and soft tissue retraction required, though this is variably reported in the literature.7,9
Establishing a standardized method in determining the risk of developing persistent postoperative dysphagia after ACDF may guide patient follow-up and management. Studies determining risk have focused on different variables without a standardized metric for risk stratification.8,10,11 Such studies have supported the individual assessment tools used in the present study to reliably evaluate dysphagia severity, though certain measures are more efficacious than others. 12 The EAT-10 assessment has often been the preferred measure of post-operative dysphagia due to its demonstrated accuracy and improved ability to identify significant dysphagia over similar methods such as the Bazaz score.12,18,26 In order to address any potential deficiencies of a single assessment and better capture the variability of dysphagia presentation and patient perception of severity, three questionnaires were used in the present study. These included the Bazaz, DSQ, and EAT-10 scores. These scores were used to evaluate an association with persistence and severity of postoperative dysphagia in patients with and without preoperative dysphagia. The results of our study demonstrate that the presence of dysphagia or pre-existing difficulties with swallowing before ACDF may lead to an exacerbation and/or prolonged persistence of dysphagia symptoms after ACDF. After further stratifying patients with immediate post-operative dysphagia into those with or without pre-existing preoperative dysphagia, our results demonstrate that patients with new-onset post-operative dysphagia perceive higher dysphagia severity. This reported difference in dysphagia severity was transient and waned by 12 weeks post-operatively. A majority of patients (56.2%) with new-onset post-operative dysphagia had complete resolution of their symptoms at the 24-week period.
Several studies have examined similar patterns of postoperative dysphagia after ACDF. At one month postoperatively, Bazaz et al. reported a 47.3% rate of dysphagia, which decreased to 20.2% at six months. 13 Separately, Tervonen et al. found an overall dysphagia incidence of 69% at the immediate postoperative visit, with 13.3% of patients reporting persistent dysphagia within a 3-9-month postoperative period. 14 These rates are comparable to our overall rates of dysphagia of 43.2% at 2-week follow-up and 18.1% at 24-weeks, providing external validity to our findings. While the majority of patients who developed new onset dysphagia following ACDF had resolution of their symptoms within the first six months postoperatively, 43.8% of patients with baseline dysphagia still experienced persistent symptoms beyond the 6-month follow-up period, demonstrating a three-fold risk increase for prolonged symptoms after surgery compared to patients without baseline dysphagia. Therefore, standard preoperative screening for dysphagia and the inclusion of dysphagia in patient selection and shared patient-surgeon decision making should be considered.
The association between operative time and incidence of postoperative dysphagia has been variably reported in the literature.4,7,15-17 Increased operative times could be attributed to the complexity of the procedure as well as other intraoperative variables that are not directly related to dysphagia status. Our study found that the operative times for patients with dysphagia were higher than in patients without dysphagia, though this difference was not statistically significant. Though the number of levels fused may also influence the operative time, evidence of this association is not consistent and was not primarily measured in our study. Regardless, surgeons should be conscientious of minimizing retractor time and pressure during surgery to avoid exacerbating symptoms in patients with underlying dysphagia.
A diagnosis of oropharyngeal dysphagia can inherently lead to necessary postoperative ENT consultation. This clinical scenario presents a challenge for ENT specialists as dysphagia-specific risk factors such as respiratory and neurological co-morbidities may not be taken into consideration during the standard pre-operative assessment for orthopedic spine surgery. 19 This may result in the development of post-operative dysphagia and associated complications such as aspiration pneumonia, dehydration, and malnutrition. 20 Although literature is limited on the dearth of ENT referrals and consults following ACDF, this trend may be attributed to reasons such as the lack of postoperative dysphagia-specific screening, failure of practitioners to recognize the severity of dysphagia, and inadequate postoperative follow up. 21 While we found that patients with new-onset post-operative dysphagia reported a higher rate of diet alteration in the immediate post-operative period, no patients in our study’s cohort required any form of postoperative-ENT consultation or assessment while in the hospital. Though the difference in postoperative ENT-related complications between those with and without an ENT assessment was not significant in our study, there may be credence to a multi-disciplinary approach to preoperative dysphagia assessment with ENT-specific variables factored in such as possible reintubation, vocal fold paralysis, and dysphonia. Given that otolaryngologic complications of anterior cervical surgery are often underreported, pre and postoperative otolaryngologic evaluations for dysphagia should be obtained for at-risk patients undergoing ACDF. 22 This is especially significant given the relatively high incidence of postoperative dysphonia and dysphagia for ACDF procedures in both our cohort and in literature.7,23
Additionally, there is a considerable number of factors to consider when regarding both immediate and long-term otolaryngology-specific complications for ACDF. Reintubation following ACDF was found to be statistically greater for older patients with multilevel fusions and, independently, those with postoperative dysphagia. 24 Regarding the latter, this is likely attributed to soft-tissue swelling following the procedure, which may result in dysphagia and ultimately reintubation. 25 Likewise, it has been reported that the risk of vocal fold motion impairment is significantly higher for lower level (C7/T1) fusions and patients who develop postoperative dysphagia or dysphonia. 23 Although early preoperative otolaryngology consultation and assessment may prove to be beneficial in reducing morbidity for this patient population in both identifying dysphagia and its associated complications, no patients in our study were reintubated, and there were no differences between groups in rates of speech pathology consultation, adjunct cervical spine MRI imaging, or other postoperative events affecting length of stay in our study.
There are a number of limitations to the present study. Although a multicenter prospective study increases the diversity of our patient population, our results were limited to patients within the included sites. The small patient cohort (n = 168) and patients lost to follow-up limit the statistical power of the study. Our loss to follow-up rate was 8.3%, 10.1% and 17.8% at 6 weeks, 12 weeks, and 24 weeks post-operatively, respectively. Specifically, with respect to patients with preoperative dysphagia, there was a higher attrition rate of 17.4% and 30.4% at 12 and 24 weeks post-operatively, which may bias our results. The late increase in dysphagia rates in the preoperative dysphagia cohort may derive from attrition between 12 and 24 week cohorts and therefore, should be interpreted cautiously. Although there were no differences in post-operative in-hospital events and outcomes such as diet alteration, postoperative steroids, speech pathology consultation, adjunct cervical spine MRI, and nasogastric tube placement, the interpretations of these outcomes are limited. The study was not powered to detect statistically significant differences in these low incidence outcomes, but rather to qualitatively comment on them. Additionally, inherent differences in the content of the three questionnaires used to stratify risk may limit the effect of using a single instrument to determine preoperative dysphagia risk. For instance, Skeppholm et al. found that Bazaz scores did not correlate with the DSQ in the context of a validation study. 11 This may affect the generalizability of our current study given that one of three instruments was administered to patients as an analogue for dysphagia severity. Despite this, our findings indicate that all three instruments were highly correlated with each other (r > .800), suggesting a low likelihood of heterogeneity between instruments. A standardized metric for quantifying dysphagia risk may provide a better understanding of dysphagia incidence rates better suited for ACDF patients.
Conclusion
Patients undergoing ACDF with pre-existing dysphagia have greater postoperative dysphagia severity and demonstrate a three-fold increased risk of prolonged post-operative dysphagia compared to patients without pre-existing dysphagia. Furthermore, patients with new-onset post-operative dysphagia perceive higher dysphagia severity in the early post-operative period compared to those with pre-existing dysphagia. This difference in dysphagia severity is transient and wanes by 12 weeks post-operatively. The majority of patients with new onset post-operative dysphagia report complete resolution of their symptoms by 24 weeks post-operatively. Patients with pre-existing dysphagia return to baseline levels of dysphagia severity after a prolonged postoperative period. There was no difference in rates of ENT consultation, length of stay, or other in-hospital events between groups. These findings elucidate the impact of pre-operative dysphagia on post-operative dysphagia duration and severity and may be used to guide perioperative counseling of patients.
Footnotes
Declaration of Conflicting Interests
Funding
IRB Approval
ORCID iDs
