Evaluating the Association of Clinical Factors With Symptomatic Recurrence of Idiopathic Subglottic Stenosis

Key Points

Question  Are there clinical factors associated with the symptomatic recurrence of idiopathic subglottic stenosis (iSGS)?

Findings  In this retrospective medical record review of 186 patients with iSGS, patients with class 1 obesity (but not class 2 or 3) showed shorter time to first symptomatic recurrence iSGS than underweight or normal-weight patients. Gastroesophageal reflux disease, hypertension, hyperlipidemia, and diabetes mellitus were associated with increasing body mass index but not with iSGS symptomatic recurrence on multivariable analysis.

Meaning  There appears to be an association between class 1 obesity and time to first symptomatic iSGS recurrence, but the mechanisms of disease progression are yet to be fully understood.

Importance  Idiopathic subglottic stenosis (iSGS) is a progressive and potentially life-threatening condition with very few targeted treatment options.

Objective  To characterize the clinical factors of patients with iSGS, including body mass index (BMI, calculated as weight in kilograms divided by height in meters squared), and evaluate their association with iSGS symptomatic recurrence.

Design, Setting, and Participants  This retrospective medical record review included 186 adult patients with iSGS treated at a single tertiary referral center between January 1, 1989, and December 31, 2015. All data analysis took place from January 1, 2018 to June 30, 2018.

Main Outcomes and Measures  The 3 BMI categories were examined for their association with iSGS recurrence. Outcome measurements included time to first symptomatic recurrence (TTFR) and recurrence-free survival (RFS). Comorbidities were recorded.

Results  Of the 186 patients in the study, 182 (98%) were women; mean (interquartile range) patient age, 49 (41-60) years. At iSGS diagnosis, 65 (35%) patients were underweight or normal weight; 45 (24%) were overweight; and 76 (41%) were obese (class 1, 2, or 3). Median BMI was 27.4. Ninety-one patients experienced TTFR at a median of 14 months. Compared with underweight or normal-weight patients, the hazard ratios for the associations of overweight, obese class 1, and obese class 2/3 patients with recurrence were 1.14 (95% CI, 0.65-1.99), 1.74 (95% CI, 1.04-2.93), and 1.04 (95% CI, 0.54-1.99), respectively. No differences in concomitant medical treatment regimens were found. While several comorbidities (gastroesophageal reflux disease, hypertension, hyperlipidemia, and diabetes mellitus) were associated with increasing BMI, they were not associated with iSGS symptomatic recurrence on multivariable analysis.

Conclusions and Relevance  Results of this retrospective review show that class 1 obesity was associated with an increased rate of iSGS symptomatic recurrence compared with underweight or normal-weight patients. This association was not seen in class 2 or class 3 obesity. Patients with class 1 obesity should be counseled about this risk to aid in the assessment and management of symptoms.

Subglottic stenosis can have many causes, including intubation trauma, autoimmune and inflammatory disorders, infectious processes, and congenital narrowing.¹ However, a certain proportion of these cases, roughly 15% to 30%, are considered to be idiopathic.^2,3 Idiopathic subglottic stenosis (iSGS) is a rare fibroinflammatory disease characterized by unprovoked narrowing of the upper airway at the level of the cricoid cartilage and upper tracheal rings causing life-altering dyspnea, stridor, and airway obstruction. Although the typical phenotype for iSGS is a middle-aged white woman, the natural history, causative factors, and pathophysiology of the disease are ill defined.

Treatment strategies for iSGS vary greatly across institutions, and opportunities for targeted treatment breakthroughs remain elusive. Symptomatic improvement, therefore, remains the guiding principle in disease treatment. Surgical interventions include both endoscopic (dilation vs mucosal-sparing wedge excision) and open procedures, with cricotracheal resection proving to be the most effective long-term treatment modality for refractory disease.^4,5 Additionally, endoscopic injection of steroids and application of chemotherapeutic agents have both been used to reduce recurrence.^6,7 Multimodal therapy has also been implemented at some institutions including the use of immunosupprresants.⁸ This includes a medical regimen targeted at treating the potential sources of inflammation, which has also been shown to decrease disease recurrence rates. These treatments include dual acid suppression therapy, high-dose inhaled corticosteroids, and daily antibiotic use.⁶

In patients with iSGS, several other comorbidities are commonly seen, raising the question about the association of these comorbidities with disease severity, progression, and recurrence. There has been some evidence suggesting an association between obesity and diabetes mellitus (DM) in adults with acquired subglottic stenosis.⁹ As obesity rates in the United States continue to rise, with nearly one-third of adult patients now considered to have obesity, the effect of obesity on disease progression, severity, and prevalence should be explored.¹⁰ The association of body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) with iSGS and its potential effect on disease severity has yet to be elucidated, but findings could drive future disease treatment and prevention. This present cohort study attempts to illustrate the association of BMI with the symptomatic recurrence of iSGS.

Following Mayo Clinic institutional review board approval (IRB 12-008100), a retrospective medical record review was performed. Patient written informed consent was waived for deidentified data. At a tertiary referral center for iSGS, the records of all patients older than 18 years who underwent treatment of iSGS between January 1, 1989, and December 31, 2015, were reviewed. All analysis took place between January 1, 2018, and June 30, 2018. The disease was identified by clinical examination using tracheoscopy, and the diagnosis of iSGS was one of exclusion. Patients with a history of prolonged, multiple (>1), or traumatic intubation were excluded. Similarly, history of tracheal trauma, prior tracheal infection, known rheumatologic disease, granulomatosis with polyangiitis, inhalational injury, or any other known causes of subglottic stenosis were also excluded in analysis. In total, 186 patients were included.

Dyspnea was the primary indication for operative intervention, which required endoscopic identification of subglottic stenosis. On occasion, a staged second procedure was completed to address higher-grade stenosis. In general, patients were treated surgically for worsening symptom severity and/or Cotton-Meyer stage 2 or 3 disease, as identified on clinical examination (tracheoscopy). Only when there appeared to be disparity between symptoms and subglottic patency, cardiopulmonary evaluations were completed prior to surgical intervention. Preoperative pulmonary function testing data were available for a majority of the patients but not for a majority of the cohort in the follow-up period of recurrence, and therefore the data set was considered incomplete, and it was not included in the analysis. General anesthesia with intermittent endotracheal ventilation or jet ventilation technique was used to facilitate access to the subglottis. Triamcinolone acetonide was injected into the subglottic scar, and mucosal wedge excision of the scar was completed using carbon dioxide laser via direct microlaryngoscopy. Following resection, mitomycin C at a concentration of 0.5 mg/mL was applied to all raw surface areas for 2.0 to 2.5 minutes, based on surgeon preference and the patient’s pulmonary reserve.

Consistent with previous reports by Gelbard et al,¹¹ symptomatic recurrence was defined as symptoms severe enough to require a repeated surgical intervention.¹¹ Time to first symptomatic recurrence (TTFR) was defined from the date of first surgical intervention to the date of the second surgery institution. Recurrence-free survival (RFS) was calculated from the date of first surgical intervention to recurrence or the last known date of follow-up. The duration of follow-up was calculated from the date of the first surgical intervention to the date of the second surgical intervention or the last known date of follow-up.

Continuous features were summarized with medians, interquartile ranges (IQRs), and ranges. Categorical features were summarized with frequency counts and percentages. The BMI categories used were those accepted by the National Institutes of Health and the World Health Organization; underweight (<18.5), normal weight (18.5-24.9), overweight (25.0-29.9), obese 1 (30.0-34.9), obese 2 (35.0-39.9), and obese 3 (≥40.0).^12,13 Recurrence-free survival was estimated using the Kaplan-Meier method. The association of BMI category with time to recurrence was evaluated using a Cox proportional hazards regression model and summarized with hazard ratios (HRs) and 95% confidence intervals (CIs). Multivariable analysis was also performed.

A summary of clinical features of the entire cohort is described in Table 1. Of the 186 patients, 98% (n = 182) identified as female. The median age at symptom onset was 46 years, with a diagnosis of iSGS being made at a median age of 49 years. The median BMI at diagnosis was 27.4 (overweight). Within the cohort, 33% (n = 62) were normal weight; 24% (n = 45) were overweight; 22% (n = 41) were obese 1; 11% (n = 21) were obese 2; and 8% (n = 14) were obese 3. A majority of patients, 72% (n = 132), had their first disease treatments at the study institution. These first treatments included mucosal-wedge sparing technique (79%; n = 142), mucosal-wedge sparing technique with dilation (10%; n = 19), airway reconstruction (7%; n = 12), dilation only (<1%; n = 1), tracheostomy (<1%; n = 2), and use of the shaver (1%; n = 1).

Clinical features of the cohort broken down by BMI category are summarized in Table 2. Nearly all patients had the symptom of dyspnea on exertion (99%; n = 177), while a minority experienced dyspnea at rest (12%; n = 22). Comorbid conditions of gastroesophageal reflex (GERD), hypertension (HTN), hyperlipidemia/dyslipidemia (HLD), and DM differed among the BMI categories. For example, when comparing normal-weight and underweight patients with overweight patients with iSGS, HLD and HTN were more prevalent in the overweight group. When comparing normal-weight and underweight patients with the obese 1 group, GERD, HLD, and HTN were more prevalent in the obese 1 group. When comparing the normal and underweight patients with the obese 2/3 group, GERD, DM, HLD, and HTN were all more prevalent in the obese 2/3 group. However, the percentage of patients with asthma or reactive airway disease was similar among the BMI categories.

In regard to RFS, 9 patients without surgical intervention and 7 patients who were not followed up past the date of the first surgical intervention were excluded from the assessment. Of the remaining 170 patients, 91 experienced their first recurrence at a median of 14 months following the first surgical intervention (IQR, 4.8-33.9 months) (Figure 1). The median duration of follow-up for the 79 patients who did not experience recurrence was 13.2 months (IQR, 3.6-43.2 months). Overall, the estimated RFS rate reached 50% at 30 months. Estimated RFS rates for all 170 patients (95% CI; n = number still at risk) at 1, 3, and 5 years following the first surgical intervention were 70% (95% CI, 63%-78%; n = 91), 47% (95% CI, 39%-57%; n = 47), and 31% (95% CI, 23%-42%; n = 18), respectively. Estimates of RFS rates based on BMI categories were also performed (Figure 2). Of those who experienced recurrence, 32 (35%) were underweight or normal weight; 20 (22%) were overweight; 26 (28%) were obese class 1; and 13 (14%) obese class 2/3 patients over a median follow-up of 1.7, 1.2, 0.5, and 1.0 years, respectively. Estimated RFS rates (95% CI; n = number still at risk) at 1, 3, and 5 years following the first surgical intervention were 76% (95% CI, 65%-88%; n = 36), 53% (95% CI, 40%-69%; n = 21), and 42% (95% CI, 29%-62%; n = 9), respectively, for underweight or normal-weight patients; 73% (95% CI, 59%-90%; n = 22), 48% (95% CI, 33%-72%; n = 10), and 28% (95% CI, 14%-55%; n = 4), respectively, for overweight patients; 55% (95% CI, 41%-75%; n = 17), 34% (95% CI, 21%-56%; n = 8), and 18% (95% CI, 7%-44%; n = 3), respectively, for obese class 1 patients; and 75% (95% CI, 59%-94%; n = 16), 51% (95% CI, 33%-80%; n = 8), and 26% (95% CI, 9%-69%; n = 2), respectively, for obese class 2/3 patients.

Obese class 1 patients experienced a HR 1.74 times that of normal or underweight patients for TTFR. Compared with underweight or normal-weight patients, the HRs for the associations of overweight, obese class 1, and obese class 2/3 with TTFR were 1.14 (95% CI, 0.65-1.99), 1.74 (95% CI, 1.04-2.93), and 1.04 (95% CI, 0.54-1.99), respectively (Figure 3). Multivariable analysis did not show any other factors associated with recurrence. Importantly, medical treatment modalities were similar among the BMI categories.

To our knowledge, the present study includes one of the largest cohorts of patients undergoing management for iSGS with review of associated comorbidities. Although originally developed for population-scale analysis, BMI can be used as a tool in the clinical setting. The BMI categories accepted by the National Institutes of Health and the World Health Organization include underweight (<18.5), normal weight (18.5-24.9), overweight (25.0-29.9), obese 1 (30.0-34.9), obese 2 (35.0-39.9), and obese 3 (≥40.0).^12,13 Currently, in the United States, approximately 39% of people are obese, which is a rate that has been increasing.¹⁰ Similarly, the present study shows an obesity rate similar to the national average: 41% of patients were obese. In regard to disease progression, patients in the obesity 1 category had a higher likelihood of recurrence (HR, 1.74) compared with normal-weight and underweight individuals as well as a shorter disease-free interval. However, this association seemed to diminish in the obesity class 2/3 category. These results are not intuitive. Perhaps obesity drives the disease process in all categories, but the symptoms for which patients seek medical attention become less bothersome or noticeable for those in higher obesity classes because they may already lead a less active lifestyle, leading to an underrepresentation of recurrences in the obesity 2/3 class in the present study despite similar follow-up.

If it is accepted that iSGS is in part due to or exacerbated by inflammation, then obesity might contribute to the underlying disease process because there is evidence to support obesity as a chronic inflammatory state. Obesity itself promotes a proinflammatory state within the body as evidenced by elevated inflammatory markers and endothelial dysfunction.^14,15 Perhaps the proinflammatory factors or endothelial dysfunction makes patients with iSGS more susceptible to inflammation and stenosis. Researching how obesity affects the microenvironment of the airway via pathologic specimens across BMI categories could explore this theory further. Similarly, there are several obesity-related abnormalities that occur concomitantly to comprise the metabolic syndrome, which has also been deemed a chronic inflammatory process.¹⁶ Metabolic syndrome encompasses several of the comorbidities also seen in patients with iSGS, including HTN, HLD, abdominal obesity, and pre-DM or DM. The effect of concomitant diseases and the role of metabolic syndrome on disease progression are out of the scope of the present study but are being further explored prospectively.

Idiopathic subglottic stenosis affects the female population almost exclusively; the present findings support this notion in that 98% of patients with iSGS were female. The role of hormones in iSGS, however, remains controversial. Three studies have suggested a possible hormonal cause, while 1 study showed the lack of estrogen receptors in disease tissue.^17-19 In regard to hormonal changes and obesity, it is thought that with increased body fat content, there is an increased level of total body estrogen by a process termed peripheral aromatization occurring in the adipose tissue.²⁰ Conceivably, in obesity, there is then an increase in total bioavailable estrogen, increasing the risk for developing or worsening progression of iSGS. The present findings demonstrate an elevated recurrence risk only in the obesity 1 category. Therefore, these data do not support an obesity-related hormonal mechanism because if they did, the RFS would be expected to decrease in a linear fashion as obesity increased.

Several comorbidities including GERD, HTN, HLD, and DM are commonly seen in association with iSGS.^9,21-25 The national prevalence of GERD, HTN, HLD, and DM are 18% to 28%, 29%, 42%, and 9.4%, respectively.^26-29 In the present iSGS, these comorbidities are seen in a prevalence of 41%, 35%, 33%, and 8.6%, respectively. However, multivariable analysis did not show any of these factors to be associated with recurrence. It is important to keep in mind that these comorbidities are also seen in those with elevated BMI without iSGS; therefore, the role of these concomitant disease processes in the development or propagation of iSGS is controversial.³⁰ This is important data, however, which highlights that iSGS is perhaps not limited to “healthy” middle-aged white women and that understanding the associated diseases and risk factors may allow for an opportunity to treat patients more comprehensively.

Limitations of this study include its retrospective nature and smaller numbers of obesity 2 and 3 categories. Additionally, it is preferred to record longer median duration of follow-up for patients who do not experience recurrence compared with the median time to first symptomatic recurrence. These points should be considered when interpreting the results. As with any rare disease, continued collaboration and/or meta-analysis would be useful in corroborating these findings.

Class 1 obesity was associated with increased recurrence rates compared with underweight or normal-weight patients, but this association was not seen in class 2 or class 3 obesity. There appears to be an association between class 1 obesity and TTFR, but the mechanisms of iSGS disease progression are yet to be fully understood. GERD, HTN, HLD, and DM are seen in high prevalence with increasing BMI and iSGS, but multivariable analysis did not show them to be associated with TTFR.

Corresponding Author: Jan L. Kasperbauer, MD, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (kasperbauer.jan@mayo.edu).

Published Online: May 9, 2019. doi:10.1001/jamaoto.2019.0707

Correction: This article was corrected on May 23, 2019, to correct a typographical error in the abstract, not affecting data.

Author Contributions: Drs Menapace, Lalich, and Kasperbauer had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Ekbom, Larson, Lalich, Kasperbauer.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Menapace, Larson, Lalich.

Critical revision of the manuscript for important intellectual content: Menapace, Ekbom, Larson, Edell, Kasperbauer.

Statistical analysis: Larson, Lalich.

Obtained funding: Menapace, Larson, Lalich, Edell.

Administrative, technical, or material support: Larson, Edell, Kasperbauer.

Supervision: Ekbom, Edell, Kasperbauer.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by the Department of Otorhinolaryngology at the Mayo Clinic Rochester, Minnesota.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: An earlier version of this work was presented at the American Academy of Otolaryngology–Head and Neck Surgery Annual Meeting; September 29 to October 2, 2013; Vancouver, British Columbia, Canada.