Taylor SC, Clayburgh DR, Rosenbaum JT, Schindler JS. Clinical Manifestations and Treatment of Idiopathic and Wegener Granulomatosis–Associated Subglottic Stenosis. JAMA Otolaryngol Head Neck Surg. 2013;139(1):76-81. doi:10.1001/jamaoto.2013.1135
Author Affiliations: School of Medicine (Mr Taylor), Department of Otolaryngology–Head and Neck Surgery (Drs Clayburgh and Schindler), and Division of Arthritis and Rheumatic Diseases, Department of Medicine (Dr Rosenbaum), Oregon Health and Science University, Portland.
Objective To compare and contrast the manifestations and surgical management of subglottic stenosis in patients with airway obstruction attributed to granulomatosis with polyangiitis (GPA), previously known as Wegener granulomatosis, and those with idiopathic subglottic stenosis (iSGS).
Design Retrospective medical chart review. Review of subglottic stenosis cases seen in the otolaryngology department of an academic medical center from 2005 through 2010. Data were obtained on disease presentation, operative management. and findings.
Setting Tertiary referral center.
Participants A total of 24 patients with iSGS and 15 patients with GPA-associated subglottic stenosis (GPA-SGS).
Results All individuals with iSGS were female, and 40% of patients with GPA-SGS were male (P < .01). Patients with iSGS tended to have a higher Myer-Cotton stenosis grade at the time of dilation than those with GPA-SGS (P = .02). Individuals with GPA-SGS were more likely to undergo tracheotomy as a result of disease-related complications than individuals with iSGS (P < .01). No patients with an open airway reconstruction in the iSGS group required follow-up mechanical dilation. In contrast, all patients with open airway reconstructions in the GPA-SGS group underwent more than 1 subsequent airway dilation (P < .01).
Conclusions While surgical utilization is the mainstay of treatment in iSGS and GPA-SGS, iSGS occurs almost exclusively in females and presents with a greater degree of stenosis at the time of endoscopic dilation. In contrast, GPA-SGS is associated with greater rates of tracheotomy. Open airway reconstruction may be used in the treatment of iSGS and GPA-SGS but is much more effective in iSGS.
Acquired subglottic stenosis (SGS) describes airway narrowing in the area of the cricoid and is associated with prolonged intubation or external trauma. Other patients acquire SGS from a systemic autoimmune etiology, such as granulomatosis with polyangiitis (GPA), previously known as Wegener granulomatosis. However, in about 20% of cases there is no identifiable precipitant, and patients are therefore classified as having idiopathic subglottic stenosis (iSGS).1 While posttraumatic SGS is often evident based on patient history, discriminating between iSGS and GPA-SGS can be diagnostically challenging, especially in cases in which SGS is the presenting symptom of GPA. Presenting symptoms of dyspnea, noisy breathing, and voice changes can occur in both groups.2,3 Anatomically, the cohorts appear different, as scar formation in patients with GPA-SGS and iSGS typically involves soft tissue (membranous stenosis), but intubation and trauma-related SGS typically leads to firm, cartilaginous scar tissue.4 GPA-SGS is diagnosed based on the presence of autoantibodies or characteristic findings on biopsy, but negative results cannot reliably rule out autoimmune-mediated SGS, especially when the disease has limited organ involvement.5,6 Indeed, given the many similarities between patients with iSGS and those with GPA-SGS, some have suggested that iSGS may represent a spectrum of autoimmune-mediated SGS that is not yet fully understood and that GPA-SGS is but a part of this autoimmune SGS spectrum.
Despite the similarities noted between iSGS and GPA-SGS, there is a paucity of data in the literature directly comparing these entities, specifically with regard to therapeutic and surgical outcomes. Given the systemic nature of GPA, it seems reasonable to assume that these patients would have more airway difficulty than those with idiopathic SGS. Furthermore, one would predict that patients with GPA would have a worse clinical response to surgical treatment, including a need for more procedures and more frequent failure of open procedure, although data to support this are lacking. To further clarify this, we undertook this study to compare the manifestations and surgical management for SGS in patients with airway obstruction attributed to GPA and those with iSGS.
Following approval by the Oregon Health and Science University institutional review board, a retrospective review was performed of medical records of patients with GPA-SGS or iSGS seen in the otolaryngology department at our institution from 2005 through 2010. To identify patients, the department billing records and operative records were queried for all encounters associated with a diagnosis of GPA, laryngotracheal stenosis, or patients who had undergone endoscopic airway dilation or open airway reconstruction. Patients were classified as having GPA if (1) they had at least 1 clinical feature, such as SGS, consistent with the disease, and they were antineutrophil cytoplasmic antibody (ANCA) positive; (2) they had a biopsy finding consistent with GPA (granulomatous inflammation, vasculitis, and/or rapidly progressive glomerular nephritis); or (3) they manifested at least 2 signs of disease (laryngotracheal involvement, septal perforation, sinonasal involvement, nasolacrimal involvement, recurrent otitis media, or characteristic renal or pulmonary involvement). Patients were classified as having iSGS if they did not have a history of laryngotracheal trauma or tracheotomy and the airway narrowing could not be attributed to another cause, such as malignant disease or a systemic autoimmune condition.
Once patients were identified, all documentation, including pre-2005 encounters, was reviewed. Data were obtained on age at diagnosis, diagnostic procedures and laboratory tests, and therapeutic management, including immunosuppressive therapy and surgical procedures. SGS was diagnosed and evaluated by flexible fiber-optic examination or by intraoperative direct laryngoscopy. Extent of laryngotracheal involvement and gross characteristics of lesions were assessed. The Myer-Cotton staging system (MCS), which was originally developed as a pediatric SGS scale but has since been implemented in monitoring adult SGS, was used to describe the stenosis based on the percentage relative reduction in cross-sectional area of the subglottis. Four grades of stenosis are described: grade 1 lesions have less than 50% obstruction, grade 2 lesions have 51% to 70% obstruction, grade 3 lesions have 71% to 99% obstruction, and grade 4 lesions have no detectable lumen or complete stenosis.7 Grade of stenosis was not documented in the operative record in 17 dilations. Airway dilations were performed with direct microlaryngoscopy using continuous radial expansion balloons (Boston Scientific), Jackson laryngeal dilators, or rigid bronchoscopic dilation. When comparing rates of surgical outcome, a minimum follow-up time of 6 months after the operative date was required for inclusion. Gastroesophageal reflux disease was diagnosed via esophagoscopy demonstrating esophagitis in 3 patients, pH probe testing in 5 patients, and clinical improvement of reflux symptoms with proton pump inhibitor therapy in 4 patients.
Descriptive statistics, t test, Fisher exact test, Mann-Whitney U test, and χ2 analysis for categorical data were performed (P < .05 denoted significance).
A total of 39 patients were identified for the study, 24 with iSGS and 15 with GPA-associated subglottic airway obstruction (Table 1). Aside from the absence of males in the iSGS group, no significant differences in patient demographics were noted. Of note, 4 patients with GPA-SGS in our cohort (27%) were diagnosed as having GPA when younger than 20 years. At the date of last follow-up, 7 patients with GPA-SGS (47%) exhibited disease involvement restricted to the head and neck while 8 (53%) had systemic involvement, including renal and/or pulmonary manifestations. The cohort was followed for 177 patient-years. The mean and median periods of follow-up for the GPA-SGS group were 8.2 and 9.9 years, respectively. In comparison, the mean and median lengths of follow-up for the iSGS group were 2.8 and 1.8 years, respectively (P < .01).
Diagnosis of the 15 patients with GPA is illustrated as follows:
The patient diagnosed by clinical features alone was male and, as is typical with GPA,8 had additional disease involving the nose and sinuses. Furthermore, although his autoantibody ANCA titers were not positive, they were interpreted as having an atypical pattern. Given these considerations, his SGS was attributed to GPA rather than an idiopathic etiology.
The severity and location of stenosis observed during dilation was assessed. Patients with iSGS were found to have significantly worse stenosis based on MCS grading than patients with GPA-SGS (Table 2). There were no significant differences in the location of stenosis seen at initial dilation, although there was a trend toward more circumferential stenoses in the patients with GPA (Table 3).
Surgical treatment of SGS consisted of both endoscopic dilation and cricotracheal resection. Endoscopic dilation technique did not vary substantially between groups, aside from less frequent use of the carbon dioxide laser in the GPA group (Table 4). Patients with GPA underwent a mean of 3.53 surgical dilations per patient compared with 2.54 in those with iSGS (P = .44). Seven patients with GPA-SGS (47%) required fewer than 2 airway dilations compared with 11 of those with iSGS (46%) (P < .99).
Definitive operative resection or reconstruction was attempted in both groups; 5 of those with GPA-SGS (33%), and 6 of those with iSGS (25%). While no patients with an open airway reconstruction in the iSGS group required follow-up mechanical dilation, all patients with open airway reconstructions in the GPA-SGS group underwent more than 1 subsequent airway dilation (P < .01). Following open airway reconstruction, 1 patient with GPA-SGS underwent subsequent tracheotomy. Open airway reconstruction led to permanent decannulation of 2 previously tracheotomy-dependent patients with GPA-SGS. Six patients with GPA-SGS (40%) underwent tracheotomy as a result of disease-related complications and 2 (13%) remained tracheotomy dependent at the date of last follow-up. No patients with iSGS required tracheotomy as a result of a disease-related complications (P < .01).
We examined the impact of various factors on the success of airway procedures within each group (Table 5 and Table 6). The presence of gastroesophageal reflux disease (GERD) and the operative use of carbon dioxide laser were not found to have an impact on the rate of surgical utilization. Male patients with GPA-SGS had a shorter time until additional procedures were needed than female patients with GPA-SGS, while the presence of a previous tracheostomy showed a nonsignificant trend toward worse outcomes. When patients from both cohorts (GPA-SGS and iSGS) were pooled into a single group, none of these factors (sex and history of tracheostomy or GERD) had any significant impact on time until additional procedures were needed.
We analyzed the utilization of systemic immunotherapy within both groups. All patients with GPA-SGS and 50% of iSGS individuals received systemic immunotherapy at some point throughout follow-up as part of disease management; all patients with iSGS received corticosteroids, while patients with GPA received a mix of corticosteroids (n = 13), methotrexate sodium (n = 11), and cyclophosphamide (n = 9). Use of immunosuppressive medication was not associated with longer procedure-free intervals.
When traumatic causes are not readily identifiable by patient history, determining the etiology of SGS can be diagnostically challenging. While many patients with nontraumatic SGS may have a systemic autoimmune condition such as GPA, many others will have an unrevealing autoimmune workup. In the absence of any identifiable cause, these patients are considered to have iSGS, although there is some speculation that this may be due to some unknown autoimmune mechanism. This study was conducted to better define the similarities and differences in presentation and therapeutic management of iSGS and GPA-SGS.
Previously, it has been demonstrated that GPA-SGS affects men and women equally.2 In contrast, iSGS almost exclusively affects women3,9,10; it is thought to predominantly affect women owing to estrogen-mediated alterations to wound-healing responses in the subglottic airway.11 Our study is consistent with these observations, further confirming the tendency for iSGS to disproportionately affect men and for GPA-SGS to affect both men and women. Interestingly, we observed that male GPA-SGS patients underwent more frequent subglottic airway surgical procedures than female patients with GPA-SGS. This may be due, in part, to previous trends noting that male patients with GPA tend to develop a more severe form of the disease than female patients with GPA.12
The median age of initial presentation in patients with GPA-SGS (36.3 years) was almost 9 years younger than that of the iSGS group (45.2 years). Although not statistically significant (P = .24), this finding is broadly in line with that of previous research, and we speculate that had our cohort been larger, the observations would have been significant. A study of 52 patients with iSGS found the average age of initial presentation to be 43.5 years,9whereas the median age at which SGS was diagnosed in a cohort of patients with GPA was 26 years.2 Furthermore, patients with GPA and SGS are frequently diagnosed as having GPA at a very young age; in fact, up to 44% are diagnosed before the age of 20 years.13 27% of patients with GPA-SGS in our cohort were diagnosed as having GPA when younger than 20 years.
GERD has been implicated in the development of SGS and has been identified as a probable precipitant of iSGS.14- 17 However, some question the existence of a direct association.9 GERD has also been explored as a possible cause of GPA-SGS, but evidence of a definitive link has yet to be identified.18 The most compelling data to date come from a study by Blumin and Johnston19 demonstrating pepsin in the larynx and trachea in 59% of patients with iSGS, but none in matched control patients. Half of our iSGS group either had a history of, or was empirically treated for, GERD, which was statistically no different from the comparison GPA-SGS group. Furthermore, the rate of surgical utilization between those with a diagnosis of GERD and those without was no different in both groups. While our results fail to demonstrate a difference in the rate of GERD and SGS in the iSGS and GPA-SGS groups, understanding the impact of GERD on the development of SGS will be best accomplished through continued prospective studies.
Operative management strategies for subglottic stenosis are focused on improving the airway, either via endoscopic dilation of the stenosis, excision of the stenosis with laryngotracheal reconstruction, or bypassing the stenosis with tracheostomy. Carbon dioxide laser resection and/or intralesional corticosteroid injection are common adjuvant treatments to endoscopic dilation. Interestingly, in our series we found that laser resection was utilized more frequently in patients with iSGS than in those with GPA-SGS. This may in part be explained by practices of the operating surgeon or a reluctance to use the carbon dioxide laser if there is a possibility of active GPA within the stenosis. While previous studies have shown the successful use of the carbon dioxide laser for GPA-SGS,20 a general principle in the treatment of GPA-SGS is to avoid as much airway manipulation as possible when active disease is present.
Intralesional corticosteroid injections at the time of manual dilation were documented in 85% of cases. Previous research of an intratracheal dilation-injection technique using glucocorticoids in GPA-SGS has shown this to be effective21 and possibly a preferred method of immunosuppressive therapy in GPA isolated strictly to the subglottis.2 While the role of intralesional corticosteroids in iSGS is less clear, it is often considered an adjunct to dilation to prolong the time between procedures. Definitive treatment of iSGS is thought to be most likely achieved with open airway reconstruction.3,10 More recently, mitomycin C has also been used as an inhibitor of fibroblastic-mediated scar formation in laryngotracheal stenosis.11,22,23 In our cohort, only 2 patients received mitomycin C; thus, it is impossible to derive any conclusions about this therapy. Future research will be needed to better define its role in the management of GPA-SGS and iSGS.
Our data demonstrated more severe stenosis as measured by MCS at the time of dilation in patients with iSGS than those with GPA-SGS, with 33% of dilations in patients with GPA-SGS and 68% of dilations in patients with iSGS classified as MCS 3. Although there is a perceived reluctance to operate on patients with GPA, our experience indicates they often undergo dilation for smaller degrees of stenosis. This may indicate that patients with GPA-SGS experience more clinically significant symptoms than those with iSGS for a given grade of stenosis. Alternatively, underlying sinonasal or pulmonary involvement with consequent increased work of breathing in individuals with GPA-SGS could explain the larger diameter airway at the time of dilation. Patients with GPA-SGS may also have longer or more irregular sections of stenosis that result in more turbulence and poorer airflow than those with iSGS with comparatively discrete and symmetric stenoses. Further research will be needed to explore the differences between the dyspnea in these 2 groups.
Although the MCS at the time of dilation was different between the 2 groups, we found identical percentages of patients with GPA-SGS and iSGS undergoing more than 1 endoscopic dilation and open airway reconstruction. Our data further indicate that iSGS cases classified as having MCS 2 have more days between mechanical airway dilations than those with GPA-SGS. A similar trend, albeit not significant, is noted in average days between dilations in patients classified as having MCS 3.
The percentage of patients with GPA-SGS undergoing open airway reconstruction in our group is similar to those of other studies,13,20 and we recently reported on the efficacy of airway reconstruction in GPA-related laryngotracheal stenosis.24 It is important to note that open airway reconstruction was much more effective for iSGS than for GPA-SGS. In patients with iSGS, open airway reconstruction could be considered definitive management, with no need for tracheostomy afterward and rare need for further airway interventions. However, in GPA-SGS, further dilation is the norm; as we have shown previously, the major benefit of open airway reconstruction for GPA-SGS is to effect decannulation.24 Forty percent of patients with GPA-SGS in our cohort required tracheotomy as part of disease treatment. This is consistent with other research demonstrating that between 41% and 52% of patients with GPA-SGS require tracheotomy.2,13,20 No patients with iSGS in our cohort required tracheotomy owing to disease-related complications, which is less than a previous study showing a 20% tracheotomy rate.25
Owing to the nature of retrospective medical chart reviews, our study has several inherent limitations. While specialists at tertiary referral centers follow patients with chronic medical conditions longitudinally for many years, the same is not true of some conditions, such as SGS, that may resolve after 1 or more treatments. The mean length of follow-up for patients with iSGS at our institution was 2.8 years compared with 8.2 years in those with GPA-SGS. This discrepancy, although informational for comparing disease chronicity between groups, does not allow for an accurate comparison of rate of surgical dilations over time. Thus, we viewed the observation that patients with GPA-SGS undergo less frequent surgical utilization as being due to loss of follow-up and attributable to the nature of retrospective reviews involving tertiary referral centers. It should be noted we also explored the possibility of systemic immunosuppressive therapy, which was used by all patients with GPA-SGS and half of those with iSGS, as an additional factor contributing to the decreased rate of surgical utilization in the GPA-SGS group. However, owing to a lack of numbers and prescribing variability between patients and procedures, the effect of immunosuppressive therapy on time between airway procedures could not be accurately assessed. The true rate of surgical utilization and systemic immunosuppressive therapy efficacy in these patients will be best determined by future prospective studies. In further considering the operative demand of both diseases, it is important to note that the median number of airway procedures in both the GPA-SGS and iSGS groups was 1 surgical intervention per patient. Thus, while both groups had individuals requiring chronic follow-up and multiple airway procedures, 1 operation provided definitive treatment for many of the patients.
When individually considering the duration of follow-up of the GPA-SGS group, the finding of an average of 0.47 surgical dilations per patient-year of follow-up could be viewed as representative for patients with GPA-SGS requiring referral to a head and neck surgeon. A similarly constructed previous study that followed patients for an average of 6.4 years demonstrated a comparable rate of surgical utilization in those with GPA-SGS: 0.36 surgical procedures per patient-year of follow-up.20
In conclusion, although several similarities exist between GPA-SGS and iSGS, iSGS occurs more often in women and presents with a greater degree of stenosis. GPA-SGS requires more long-term management and is associated with a higher rate of tracheotomy. While open airway reconstruction may be used in both iSGS and GPA-SGS, it is much more effective in iSGS. Within the GPA-SGS group, the rate of surgical utilization in individuals with GPA-SGS requiring tracheotomy was not significantly different from those whose disease did not require tracheotomy and male patients with GPA-SGS required more frequent subglottic airway procedures than female patients with GPA-SGS.
Correspondence: Joshua S. Schindler, MD, Department of Otolaryngology–Head and Neck Surgery, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Mail Code PV-01, Portland, OR 97239 (firstname.lastname@example.org).
Submitted for Publication: July 13, 2012; final revision received September 27, 2012; accepted October 15, 2012.
Author Contributions: All authors 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: Taylor, Clayburgh, Rosenbaum, and Schindler. Acquisition of data: Taylor, Clayburgh, Rosenbaum, and Schindler. Analysis and interpretation of data: Taylor, Clayburgh, Rosenbaum, and Schindler. Drafting of the manuscript: Taylor and Clayburgh. Critical revision of the manuscript for important intellectual content: Clayburgh, Rosenbaum, and Schindler. Statistical analysis: Taylor and Clayburgh. Study supervision: Rosenbaum and Schindler.
Conflict of Interest Disclosures: Dr Rosenbaum is a paid consultant for Genentech.
Funding/Support: Research by Dr Rosenbaum is, in part, funded by Research to Prevent Blindness (New York), Genentech, the Stan and Madelle Rosenfeld Family Trust, the William and Mary Bauman Foundation, and the William C. Kuzell Foundation.
Previous Presentation: This study was presented at the Annual American Broncho-Esophagological Association Meeting; April 19, 2012; San Diego, California.
Additional Contributions: Jess Mace, MPH, provided assistance with the statistical analysis.