Figure 1. Forest plot of included studies demonstrating the relative risk ratio for surgical failure of supraglottoplasty in patients with associated comorbidities vs those with isolated laryngomalacia. A risk ratio greater than 1 indicates increased failure probability in those with associated comorbidities; the horizontal line, the 95% CI; and the size of the box, the sample size.
Figure 2. Forest plot of included studies demonstrating the relative risk ratio for aspiration after supraglottoplasty in patients with associated comorbidities vs those with isolated laryngomalacia. A risk ratio greater than 1 indicates increased aspiration probability in those with associated comorbidities; the horizontal line, the 95% CI; and the size of the box, the sample size.
Preciado D, Zalzal G. A Systematic Review of Supraglottoplasty Outcomes. Arch Otolaryngol Head Neck Surg. 2012;138(8):718-721. doi:10.1001/archoto.2012.1251
Author Affiliations: Division of Otolaryngology, Children's National Medical Center, Washington, DC.
Objectives To analyze the available published data on supraglottoplasty, epiglottoplasty, and laryngomalacia and to evaluate the relative risk of supraglottoplasty failure.
Design Systematic review with determination of relative risk.
Main Outcome Measures A PubMed search was performed with the following inclusion criteria: English language, human subjects, supraglottoplasty, epiglottoplasty, and laryngomalacia. The results of the included studies were summarized and analyzed. Subgroup analysis was then performed.
Results Twelve studies were identified, with 8 meeting the inclusion criteria. The overall risk ratio of surgical failure among patients with associated comorbidities compared with those with isolated laryngomalacia was 7.14 (k = 6 studies; 95% CI, 3.73-13.74; P < .001). The risk ratio for persistent or significant aspiration after supraglottoplasty among patients with associated comorbidities compared with those with isolated laryngomalacia was 4.33 (k = 3 studies; 95% CI, 1.25-15.06; P = .02). Insufficient data were available to assess outcome by age at surgery or specific technique used.
Conclusions The relative risk of supraglottoplasty failure is significantly higher among patients with associated medical comorbidities. This aggregate finding should be taken into account when parents are counseled as to the expected surgical outcome of infants with laryngomalacia who are undergoing supraglottoplasty.
Laryngomalacia is the most common congenital laryngeal cause of stridor, representing the diagnosis in approximately 45% to 75% of pediatric stridor cases. Surgical correction is reserved for severe cases that are accompanied by apneic spells, failure to thrive, and feeding difficulties, which involve fewer than 10% of patients with the condition. Endoscopic supraglottoplasty is the current mainstay and first-line operation for infants with severe laryngomalacia. This procedure was first described by Zalzal et al1 in 1987 using cold instruments. Subsequently, the carbon dioxide laser became popular, and since then, the microdebrider has been introduced.2,3 Recent studies suggest that there is no difference in outcome between the 2 endoscopic instrumentation (laser vs cold steel)4,5 techniques.
Fortunately, supraglottoplasty is remarkably successful in improving or resolving the symptoms and signs of laryngomalacia in most patients. Reported success rates range from 53% to 95%.1,6- 9 While the details on the definition vary from study to study, the success of supraglottoplasty rests on an improvement in respiratory status and feeding problems. Supraglottoplasty failures may be attributable to 1 of 2 main issues: (1) failure to resect enough supraglottic tissue or (2) comorbid conditions that contribute to the poor clinical picture. Complications resulting from supraglottoplasty are also rare. In most series, the rate is below 10%. The most common postoperative long-term complication is aspiration, with others such as granuloma formation and supraglottic stenosis being quite rare. There also appears to be no difference in complications according to the endoscopic instrumentation used.3- 5,10 However, given the low rate of expected complications, most of these case series are underpowered to identify possible differences by patient group or between techniques.
The goals of this study were to combine the reported outcome literature on supraglottoplasty and to perform a systematic review of supraglottoplasty outcomes. In doing so, we aimed to determine the relative risk of failure of the procedure from the available published aggregate data.
PubMed was searched for multiple specific search terms. The search period was from January 2001 to February 2012. No other databases were included. The specific search terms used were supraglottoplasty, epiglottoplasty, and laryngomalacia. The following inclusion criteria were then applied: English language, human subjects, and outcome data. No age criteria were applied. Studies that directly reported on the failure of surgery were included, regardless of the surgical technique that was used. Articles that did not include discreet metrics were excluded (no mention of number of success or failures or patient categorization by associated comorbidities). As mentioned, multiple supraglottoplasty techniques (microdebrider, laser, cold steel) were included, as the goal was to assess the risk of supraglottoplasty failure rather than the instrumentation used to perform it. The results of these studies were summarized in an evidence table and analyzed, with primary outcome measures of surgical success and rate of chronic aspiration as reported by each article. Surgical failure was defined as the need for revision surgery, tracheostomy tube placement, or gastrostomy. Subgroup analysis was then performed with the same outcomes comparing patients with significant associated comorbidities with those without associated comorbidities. Significant comorbidities were extrapolated from each study included and were defined as cardiac, neurologic, or gastrointestinal. All studies included a minimum of 10 patients. No prospective controlled studies were found in the literature. Only case-control and retrospective case series studies were identified and included in this analysis. Statistical analysis was performed with statistical software (MIX 2.0 PRO, Version 126.96.36.199; BioStatXl). Random-effects modeling was used to calculate summary effect measures (risk ratio) with corresponding 95% confidence intervals, and Forest plots were generated. P < .05 was considered significant.
Using the aforementioned search criteria, 12 articles were initially identified. Eight studies met the inclusion criteria (Table).3- 5,10- 14 All identified studies were retrospective in nature (retrospective case-control studies, level 4). The mean number of patients in each study was 60.1 (range, 23-136), and the mean age of the patients at the time of supraglottoplasty (in studies in which it was annotated, k = 6) was 124.9 days (range, 69-182 days). The average length of follow-up after supraglottoplasty was 8 months (k = 3 studies). Of the 8 studies identified in the literature that met the inclusion criteria listed in the Table, only 6 had sufficient information to extract comparable measures on the primary outcome of surgical success by the presence of associated comorbidities. Eustaquio et al12 presented data on aspiration risk only, without specific information as to “surgical success” by associated comorbidity. Similarly, Rastatter et al5 presented data relative to aspiration risk but not to surgical success by associated comorbidity. Only 3 studies had sufficient information to extract comparable measures on the primary outcome of aspiration risk by presence of comorbidities.
Random-effects modeling was performed to estimate surgical success rates comparing patients with isolated laryngomalacia with those with associated comorbidities. The overall risk ratio of surgical failure among patients with significant associated comorbidities compared with those with isolated laryngomalacia was 7.14 (k = 6 studies; 95% CI, 3.73-13.74; P < .001) (Figure 1), significantly demonstrating the risk of surgical failure in patients with associated comorbidities. The risk ratio for persistent or significant aspiration after supraglottoplasty among patients with associated comorbidities compared with those with isolated laryngomalacia was 4.33 (k = 3 studies; 95% CI, 1.25-15.06; P = .02) (Figure 2), also demonstrating a significant rate of persistent or new aspiration in patients with associated comorbidities. There appeared to be no difference in outcome by surgical technique or by age at surgery, although there were insufficient data reported to calculate the relative risk by those variables.
Laryngomalacia is the most frequent cause of congenital stridor, warranting surgical intervention in up to approximately 10% of children with the condition. In these instances, endoscopic supraglottoplasty is the surgical approach of choice, with good outcomes expected in most cases. Case series demonstrate little difference in outcomes or complications among the different endoscopic instruments used for supraglottoplasty. However, children with associated congenital anomalies, such as concomitant neurologic disease,4,15 cardiac disease,13 and severe gastroesophageal reflux,10,14 have been reported to be at increased risk of surgical failure, requiring higher rates of revision surgery, tracheotomy, or feeding tube insertion. The exact mechanisms for why these children seem to be at increased risk after surgery are unknown but are likely attributable to multifactorial reasons. Complex hypotonia, increased work of breathing, distorted central cardiopulmonary function, and laryngeal edema are all likely to contribute to varying degrees in individual patients. Day et al,11 in a multivariate analysis of 74 patients, reported that the only variable that contributed to supraglottoplasty surgical failure was a history of prematurity, suggesting that failure previously attributed to associated congenital anomalies may be explained by a higher rate of prematurity in these complex comorbidity cases. Along the same lines, Hoff et al13 found that infants requiring supraglottoplasty in their first 2 months of life incurred a significantly higher rate of surgical failure and revision surgery. For the purposes of categorizing patients and analyzing outcome in this review, we defined significant comorbidities as cases involving congenital cardiac disease, neurologic compromise, or severe gastrointestinal reflux. None of the studies had objective measures of cardiac or neurologic compromise degree or of reflux severity defined in their data but simply reported the number of patients in each of these categories. In general, these categories of comorbidities were grouped by the reports to evaluate whether they affected surgical outcome. Only 3 studies11,13,14 had sufficient data to determine outcome by each individual comorbidity separately; therefore, for the purposes of our analysis, all 3 comorbidities were grouped together.
The overall goal of this review was to group the recent reports on supraglottoplasty outcome to determine the relative risk of surgical failure in children with laryngomalacia. In our review, we analyzed data on 481 grouped patients. We found a relative risk of 7.14 (P < .001) for supraglottoplasty failure in infants with associated comorbidities compared with those without. This relative risk ratio signifies that, overall, the percent probability of the need for revision surgery, tracheotomy, or gastrostomy tube placement was 7.14-fold higher in children with associated cardiac anomalies than in those with isolated laryngomalacia. Furthermore, the relative risk of aspiration in these complex comorbidity cases was also found to be significantly higher at 4.33. However, we were able to include only 3 reports with sufficient data to extrapolate enough information about aspiration outcome.3,12,14 In our review, we were not able to separate the relative risk by specific comorbidity owing to the disparity in categorization in each report that we reviewed. Similarly, because of inadequate substratification in the literature, we were unable to determine the outcome of supraglottoplasty by the surgical technique used or by the severity of laryngomalacia. Therefore, it is unclear from the published medical evidence whether laryngomalacia severity alone would portend a higher risk of supraglottoplasty surgical failure.
Potential long-term complications related to supraglottoplasty include supraglottic stenosis, laryngeal cartilage damage, airway fires, granuloma formation, and death. In our review, most of the reported complications included persistent laryngomalacia, airway distress, and need for tracheotomy and were included in the surgical failure rates reported. Denoyelle et al,4 in the largest series included in this review (136 patients), reported the development of supraglottic stenosis in 5 cases (3.7%). Other studies included in this review did not report the incidence of this potential complication. Our study has several limitations. Overall, the level of medical evidence in the literature is low (level 4), as each study identified and included in our review was retrospective in nature and as such does not account for potential selection bias. The exact definition of significant associated comorbidity may have varied from study to study, and, for our analysis, we had to accept as objective the definition of each individual study. Also, the definitions of surgical failure varied from report to report and were not uniform. In 4 reports, surgical failure was explicitly stated and measured,4,10,11,13 while in 2 reports, this was inferred from the reported outcomes.3,14 To be as objective as possible, we used the number of events of revision surgery, tracheotomy tube placement, or gastrostomy tube placement in all instances to determine surgical failure. Finally, the follow-up rate for most of the studies was not reported.
In conclusion, after analyzing the medical literature on supraglottoplasty outcomes in aggregate, we can report that the results of this procedure are excellent overall and that severe complications, such as supraglottic stenosis and aspiration, are uncommon. The relative risk of revision surgery, tracheotomy tube placement, or gastrostomy is significantly higher among patients with associated medical comorbidities. This increased risk should be taken into account when counseling parents as to the expected surgical outcome of infants undergoing supraglottoplasty. Randomized, prospective studies with well-defined protocols appear to be necessary to further stratify the risk of failure by the specific medical comorbidity and/or surgical technique used.
Correspondence: Diego Preciado, MD, PhD, Division of Otolaryngology, Children's National Medical Center, 111 Michigan Ave NW, Washington, DC 20010 (email@example.com).
Submitted for Publication: April 8, 2012; final revision received May 3, 2012; accepted May 17, 2012.
Published Online: July 16, 2012. doi:10.1001/archoto.2012.1251
Author Contributions: Dr Preciado had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Preciado. Acquisition of data: Preciado. Analysis and interpretation of data: Preciado and Zalzal. Drafting of the manuscript: Preciado. Critical revision of the manuscript for important intellectual content: Preciado and Zalzal. Statistical analysis: Preciado. Administrative, technical, and material support: Preciado. Study supervision: Zalzal.
Financial Disclosure: None reported.
Previous Presentation: This study was presented in part at the European Society of Pediatric Otolaryngology Meeting; May 19-21, 2012; Amsterdam, the Netherlands.