xRead - Swallowing Disorders in the Adult Patient (October 2024)

1771

S. I. Dhar et al.: Does Medialization Improve Swallowing Function in Patients

hyaluronic acid, calcium hydroxyapatite, fat, Silastic, Teflon, Gore-Tex, Proplast), timing of procedure, type of swallowing assessment (i.e., self-reported scale, clinician reported/symptom assessment, clinical screening tool (e.g., Yale Swallow Protocol [15]), VFSS, FEES, or high resolution manometry [16–18]), timing of pre- and post intervention dysphagia assessment, and need for speech language pathology (SLP) directed swallowing therapy post-intervention. Both statistical and clinical significance of dysphagia outcomes were extracted. Each study also underwent risk of bias assessment using the rules established by the Cochrane Collaboration [19]. Studies were judged on 7 risk parameters (study rationale, study objectives, selection criteria, study replication, sam pling, detection bias, attrition bias, avoidance, and report ing bias) to have either low, unknown, or high risk of bias.

07/12/2021. Note that PubMed contains MEDLINE and that the Cochrane Library contains the Cochrane Register of Controlled Trials (CENTRAL). Additionally, a search was completed in the ClinicalTrials.gov trial registry. The searches contained three concepts: vocal fold paralysis, vocal fold medialization, and swallowing outcomes. Con trolled vocabulary, such as Medical Subject Headings and Emtree, was used where appropriate, in combination with free text keywords. No published search filters were used, and no limits were applied. The search was peer-reviewed by another medical librarian. There were 6503 total results, 3615 duplicates removed, and 2888 remaining results, which were then uploaded to the Covidence platform for title and abstract screening [13]. Zotero was used for citation manage ment [14]. The search strategies for every platform can be found in the Appendix. Five authors (A.C.D., E.J., K.M.P., M.A.R, and S.I.D) independently screened the title and abstract of identified studies using Covidence software (www.c​ovide​nce.o​ rg, Mel bourne, VIC, Australia). Two authors (S.I.D and M.A.R) then independently reviewed the full text of the remaining studies. All disagreements were resolved by a senior author (L.M.A). Study authors were directly contacted via email for missing information as needed. Prospective studies that met all criteria were then subjected to data extraction, risk of bias assessment, and outcomes analysis. The search update conducted on 7/12/2021 had a data range of 1/1/2020-7/12/2021. The same five authors per formed the title and abstract screening, and full-text review was conducted by two authors (S.I.D and L.M.A). Disagree ments were resolved by senior author (M.B.B). Data extraction and risk of bias assessment for each study meeting criteria for final review were completed indepen dently by two authors (S.I.D and M.A.R.) with data entry into a Qualtrics (2019) online form. The data were then downloaded and reviewed by a third author (M.B.B.) for completeness and disagreements. Disagreements were identified and settled with discussion and refereed by a fourth author (L.M.A). Data extracted included study date, country of first author, funding sources (government organization, association, institutional commercial, other), study type (case series, cohort, or other type), study N, study patient demographics (age, sex), cause of UVFP, method of identifying the paralysis (flexible laryngos copy with stroboscopy, flexible laryngoscopy without stroboscopy, rigid laryngoscopy, laryngoscopy not other wise specified, during FEES, laryngeal mirror, or other), type of medialization procedure (injection augmentation, transcervical medialization implant, transoral medializa tion implant) and material used (carboxymethylcellulose, Data Extraction and Risk of Bias Assessment

Statistical Methods

Descriptive statistics including proportions, means, stand ard deviations, medians, and interquartile ranges were calculated using Microsoft Excel 2019. A meta-analysis was not pursued due to substantial heterogeneity in study methods and reporting for included studies.

Results

Study Selection

We identified 2888 studies from the electronic database search and 26 additional studies from the gray literature (Fig. 1). After removing duplicates, we reviewed the titles and abstracts of 2786 studies and included 142 for full text review. On full-text review, we excluded 133 studies because of either wrong study type or study methods. The 9 remaining studies were reviewed for data extraction and risk of bias assessment; these studies contained a total of 157 patients for qualitative analysis (Table 1). The 9 studies included 7 prospective case series [20–28] and 2 prospective cohort studies [20, 23]. The studies came from 5 different countries and 1 required translation from Japa nese into English for review [26]. The 7/12/2021 search update resulted in 301 articles from which 54 duplicates were removed, leaving 247 refer ences for independent title and abstract screen. Fourteen articles met criteria for full-text review. All 14 articles were excluded due to either having an excluded article type (e.g., retrospective design) or excluded methods (e.g., no swallowing evaluation).

1 3