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

10976817, 2023, 1, Downloaded from https://aao-hnsfjournals.onlinelibrary.wiley.com/doi/10.1177/01945998211072832 by University Of Miami Libraries, Wiley Online Library on [01/07/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

3 17

4 Coulter et al

Otolaryngology–Head and Neck Surgery

regained function was excluded from the quantitative analy sis. If the original publication did not specify success of the intervention based on improvement or complete resolution of symptoms of dysphagia and/or aspiration, quantifiable out come measures were utilized to make that determination on a case-by-case basis given the available literature from the spe cific instrument used (if available) and by criteria discussed and agreed on by the authors. Liberation from the need for a gastrostomy tube or advancement of diet from NPO status (nothing by mouth) following a vocal fold medialization pro cedure was utilized as an indication of treatment success. Observation of improvement or resolution of penetration or aspiration based on FEES or VFSS was also considered treat ment success. If success was cited as a percentage of the cohort and stratified by symptoms of dysphagia and aspiration and if individual data were not available to determine overlap between these groups, the larger of the groups was considered the dysphagia group for the quantitative analysis. Studies that reported pooled outcomes that could not be individually extracted for subjects could not be quantitatively analyzed, since these results could not be dichotomized into successful or not on a patient level. In addition, studies were excluded from the quantitative analysis if outcomes of subjects who underwent IL or LF surgery were presented in the same study and could not be stratified by procedure. All studies excluded from the quantitative analysis were, however, in the qualita tive analysis. Statistical Analysis Quantitative analyses in studies with extractable data regard ing treatment success and complications were performed by compiling extracted subject-level data and using a random effects model given the variation in the source data. Specific analyses and forest plots were derived for procedure-specific success rates as well as major and minor complication rates. Standard error was estimated as the inverse of sample size. All analyses were performed in STATA IC version 11 (Stata CorpLP). Results An overall 416 publications were screened, and 26 met inclu sion criteria, as displayed in Figure1 . The studies comprised 959 patients with UVFI who underwent 916 vocal fold media lization procedures. Of these procedures, 547 were IL, 357 were LF, and 12 were LR procedures. Table 1 reviews each study, the complications, and the MINORS score for risk of bias. MINORS scores ranged from 7 to 13, with 7 of 26 (26.9%) studies regarded as having a low risk of bias. Quantitative Analysis of Complications Thirteen studies had sufficient data to be in the quantitative analysis of complications. 14,17,20,23,25,31-38 The overall esti mated complication rate (major and minor) was 10% (95% CI, 6%,14%). When stratified by procedure, the pooled esti mated complication rate (major and minor) was 7% (95% CI, 2%-13%) for IL ( Figure2 ) and 15% (95% CI, 10%-20%) for LF surgery ( Figure 3 ). When stratified by major and minor

complications, the pooled estimates of the rate of major com plications were 1% (95% CI, 0%-3%) and 2% (95% CI, 0%- 4%) for IL and LF, respectively. The minor complication rate was 5% (95% CI, 0%-10%) for IL and 14% (95% CI, 8%- 19%) for LF. Confidence intervals overlap for all estimates, suggesting no statistical significance in the complication rates between these procedures. Quantitative Analysis of Treatment Success Eleven studies had sufficient data to be in the quantitative analysis of treatment success, which encompassed 293 sub jects/medialization procedures. 19,25,27,31-35,39-41 Only studies that reported subject-level outcome data stratified by proce dure were included. An analysis of dysphagia outcomes from LR procedures could not be performed because just 7 subjects with dysphagia underwent LR in the studies. Therefore, this analysis examined outcomes for IL and LF procedures. No studies utilizing the Eating Assessment Tool–10 (EAT-10), Penetration-Aspiration Scale (PAS), or Functional Outcome Swallowing Scale met inclusion criteria for the quantitative analysis. The Functional Oral Intake Scale (FOIS) was uti lized in studies that met inclusion criteria for this analysis. Based on this scale, success was defined by liberation of the need for gastrostomy tube or resumption of a normal diet from the need for rheologic modification prior to the proce dure. One additional study utilized an instrument referred to as ‘‘dysphagia grade according to Fujishima,’’ 27 which was renamed the Food Intake Level Scale in a 2012 validation study 42 and found to be concordant with the FOIS. Like the FOIS, success was defined by liberation of the need for gas trostomy tube or resumption of a normal diet from the need for rheologic modification prior to the procedure. Based on these definitions of success, patients who underwent IL demonstrated a success rate estimate of 90% (95% CI, 75%-100%; Figure 4 ), and patients who underwent LF sur gery had a similar success rate at 92% (95% CI, 87%-97%; Figure5 ). Qualitative Analysis As detailed in Table1 , there was overall improvement in sub jective and objective measures of dysphagia in nearly all 26 studies. Rates of improvement range from 20% to 100% depending on the definition of improvement and the method of assessment. Lower success rates were cited when patient reported outcomes were noted, 12 while the highest rates are from studies with outcomes defined by oral feeding status. Studies of LF procedures cited rates of diet liberation ranging from 69% to 75%, 19,22 while studies of IL had rates ranging from 70% to 100%. 34,36,37,40,41 Patient-reported dysphagia outcome measures were more variable for LF and IL. In stud ies that attempted to compare outcomes between LF and IL procedures, 12,23,43 there were in general no significant differ ences in outcomes shortly following surgery. One would expect LF surgery to provide more durable benefit, but very few of these studies tracked outcome data for longer than sev eral months after the procedure. There were 2 studies of LR, with both indicating improvement in dysphagia over time.