Delayed Diagnosis of Idiopathic Subglottic Stenosis

Abstract

Objective:

Idiopathic subglottic stenosis (iSGS) is a rare disease, causing life-threatening dyspnea secondary to scarring. Perhaps because it is rarely encountered, there is often a delay in diagnosing iSGS. The objective of this study is to characterize diagnostic delay of iSGS, factors that prolong delay, and its impact on iSGS patients.

Methods:

A retrospective chart review of 124 iSGS patients was performed. Times of symptom onset, presentation to otolaryngologist, diagnosis, imaging, pulmonary function testing (PFTs), surgeries, emergency department (ED) visits, and hospitalizations were recorded and univariate analyses were used to identify risk factors for delay.

Results:

The median total time to diagnosis from symptom onset was 24.5 months, with time to first presentation of 6.3 months and healthcare delay of 17.8 months. 54.8% of patients were diagnosed with asthma. Earlier presentation to otolaryngologist was associated with shorter healthcare delay and total time to diagnosis (rho=0.75,rho=0.99,p<0.0001). Earlier CT imaging was correlated to shorter healthcare delay (rho=0.84,p<0.0001) and total time to diagnosis (rho=0.74,p<0.001), while earlier PFTs were correlated to shorter total time to diagnosis alone (rho=0.71,p=0.01). During evaluation, 10.5%(n=17/124) of patients had ED visits and 13.7%(n=13/124) patients were hospitalized. Before diagnosis, 7%(9/124) of patients underwent surgeries (including 3%(n=4) undergoing tracheostomy) and 8%(n=10) of patients required unplanned urgent endoscopic surgery that may have been avoided with earlier diagnosis.

Conclusion:

iSGS diagnosis is frequently delayed, resulting in additional surgeries (including tracheostomy), ED visits, and hospitalizations. Further, patients’ symptoms are commonly attributed to asthma. Earlier otolaryngologist evaluation, PFTs, and CT imaging may expedite iSGS diagnosis.

Level of Evidence:

4

INTRODUCTION

Idiopathic subglottic stenosis (iSGS) is a rare life-threatening disease, caused by extrathoracic obstruction of the subglottic and proximal tracheal airway. iSGS arises in healthy individuals, and predominantly affects women ages 30 to 50 years old, with an estimated incidence of 1:400,000.^1,2 Patients with iSGS present with respiratory symptoms consisting of progressively worsening dyspnea, accompanied by stridor and, at times, hoarseness in more advanced disease.^3,4 Individuals with this disease often require several surgeries per year to relieve symptoms, including endoscopic balloon dilation of the stenosis or resection with end-to-end anastomosis.^5,6

Due to its rare prevalence and presentation in healthy individuals, iSGS is often challenging to correctly diagnose, which leads to delay in treatment. iSGS diagnosis is frequently confounded with other diagnoses, including asthma, sinonasal allergic disease, paradoxical vocal fold dysfunction, and GERD. In prior studies, 33% to 37% of patients with iSGS were misdiagnosed with asthma prior to accurate diagnosis with iSGS.^5,7 In a survey study of patients with iSGS, 58% iSGS patients reported a delay in diagnosis.⁸ This results in consequences to individuals who have unnecessary medications or surgeries, additional physician visits or hospitalizations, and continue to suffer from dyspnea.

While patients have reported delay in diagnosis of iSGS, no studies have quantified the delay and its impact on iSGS patients. We hypothesized that diagnostic delay in iSGS is present, that there is a longer delay in patients with a previous asthma diagnosis, and that delay impacts dilation interval, with earlier referral being associated with earlier diagnosis and treatment to relieve dyspnea. Further, we hypothesized that the delay in diagnosis is due to healthcare-related delays in extended evaluation prior to accurate diagnosis to a greater extent than patient delay in reporting symptoms to their physician. The objective of this study is to assess a single institutional cohort to answer these questions, evaluate risk factors for prolonged delay, and identify the consequences of such delay on patients.

METHODS

Patient cohort

This study was approved by the Johns Hopkins University institutional review board. Patient records were reviewed for patients greater than 18 years old diagnosed with iSGS and first seen within The Department of Otolaryngology – Head and Neck surgery between 1998 and 2020 were included. Individuals were identified with ICD-10 codes for laryngotracheal stenosis (J39.8, J38.6). Patients with iatrogenic, autoimmune, or other etiologies of laryngotracheal stenosis (n=3) were excluded from the study. Autoimmune SGS was defined by a positive specific auto-antibody (i.e. c-ANCA for GPA) or SGS associated with other signs of vasculitic disorders. Patients that test negative for autoimmune labs and have no other symptoms or signs of systemic vasculitic disorders were classified as idiopathic.¹

Data collection

A retrospective chart review was performed including demographics, comorbidities, diagnostic workup, and surgeries. Diagnostic timepoints were recorded, including date of symptom onset, first presentation to otolaryngology-head and neck surgery, date of iSGS diagnosis, and date of first surgery to treat iSGS. When only the month or year was indicated in a clinical note, dates were approximated to the first day of that month or year, respectively. Total time to diagnosis was calculated as the time from symptom onset to iSGS diagnosis for patients with available data and included both patient and healthcare delay.⁹ Time to first presentation was defined as the date of symptom onset to date of first presentation to healthcare professional. On the other hand, healthcare delay was defined as the date of first presentation to healthcare professional to the date of iSGS diagnosis. Evaluation events including radiographic imaging and laboratory tests, specialist referrals, other diagnoses attributed to the symptom complaints, as well as emergency department (ED) visits, and hospitalizations attributed to symptom complaints that occurred prior to iSGS diagnosis were recorded. Among patients who underwent CT neck scans or pulmonary function testing (PFT), these dates were recorded when available in the medical record. Body mass index, comorbidities, and Charleston Comorbidity Index were recorded using the patient’s first office visit.

Surgical data

Cotton-Myer grade was assessed at initial dilation procedure. When the degree of stenosis was described by airway diameter alone, the grade was estimated by computing the ratio of measured diameter to average diameter of the subglottis (17mm in men and 14mm in women).¹⁰ A dilation procedure was defined as a suspension microlaryngoscopy with endoscopic dilation of the airway using either a balloon or rigid dilator.

Statistical Analysis

All analyses were conducted with STATA SE software (v16.1 StataCorp, College Station, Tx, USA) with significance level set at alpha of 0.05. Patients with diagnostic delay less than and greater than 2 years, and evaluation of less than and greater than 6 months were compared using chi-squared test for parametric variables and t-tests for continuous variables. Pearson coefficient was calculated for continuous variables and delay times including: time to first presentation, healthcare delay, and total time to diagnosis.

RESULTS

Study Cohort

Diagnostic timepoints and clinical information were reviewed for 124 patients with iSGS. The majority of patients were female (98%) with a mean age at diagnosis of 48.9 years (Table 1). Most patients had few comorbidities, with an average Charleston Comorbidity Index of 1.13. At first dilation procedure at our institution, patients had an average Cotton Myer of 2.1, which was evenly distributed between scores I, II, and III. Cotton Myer stage at first presentation to treating medical center was not associated with increased time to first presentation, healthcare delay or total time to diagnosis.

Table 1:

Characteristics of patient cohort diagnosed with idiopathic subglottic stenosis.

Patient Characteristic	Mean (SD) or % patients
Female Sex	99.1%
Age at Diagnosis (n=118)	48.86 ± 13.7 years
Race/Ethnicity (n=124)    Caucasian    African American    Asian    Hispanic/Latino	96.8% 0.08% 1.6% 0.08%
Insurance Coverage (n=124)    Commercial    Medicare/Medicaid    Other*	75.2%     16%     8.8%
BMI (n=121)	29.6 ± 8.5 kg/m2
Former Tobacco Use (n=124)	7.2%
Cotton Myer (n=93)	2.1 ± 0.85
I    II    III    IV	31.2%     26.9%     41.9%     0%
Charleston Comorbidity Index (n=120)	1.13 ± 1.38

^*Other insurance coverage includes Tricare, HMO, MCO, and None

Delay Intervals

Figure 1 shows the diagnostic timeline including a patient’s median time from symptom onset to: first health care provider presentation, first otolaryngology presentation, date of iSGS diagnosis, and first surgery to treat iSGS. The median total time to diagnosis (time from symptom onset to iSGS diagnosis) was 24.5 months, with a time to first presentation of 6.3 months and a healthcare delay of 17.8 months. A total time to diagnosis of greater than 5 years was present in 19.2% of patients. Earlier presentation to an otolaryngologist was associated with shorter time to first presentation, a shorter healthcare delay and total time to diagnosis (rho=0.59, rho=0.75, rho=0.99, respectively with p<0.0001). In contrast, earlier presentation to pulmonologist was not associated with shorter delay times. The median time from otolaryngologist referral to first dilation procedure was 1.1 months. Earlier imaging was correlated to shorter workup (rho=0.84, p<0.0001) and shorter total time to diagnosis (rho=0.74, p<0.001), while patients who underwent pulmonary function testing earlier had a shorter total time to diagnosis alone (rho=0.71, p=0.01).

Figure 1: Timeline from patients’ symptoms onset until iSGS diagnosis and treatment.

Median timepoints used to calculate the total time to diagnosis. Total time to diagnosis was divided into time to first presentation and healthcare delay (time from first presentation to iSGS diagnosis).

Referral Pattern and Evaluation

The majority of patients (57%) with symptoms of iSGS presented first to their primary care physician, with 19.6% of patients presenting to either the ED or admitted as an inpatient for further workup (Figure 2a). Symptoms at time of presentation ranged from dyspnea at rest (67%, 83/124), dyspnea on exertion (17%, 21/124), stridor (23%, 28/124), cough (17%, 21/124), dysphonia (7%, 9/124) , and dysphagia (2%, 2/124). One-way ANOVA indicated that patients with different presenting symptoms did not have different time to first presentation, healthcare delay, or total time to diagnosis. Patients saw a median of 2 specialists prior to iSGS diagnosis, with 36.3% of patients evaluated by ≥3 specialists. Patients whose evaluation took longer than 6 months saw a greater number of specialists (p=0.02). Otolaryngologists were the primary specialist to diagnose iSGS (Table 2). Patients saw a number of different specialists prior to correct iSGS diagnosis (Figure 2b) and underwent evaluation ranging from radiographic imaging, to cardiac (echocardiogram, stress testing) and gastrointestinal testing (esophageal manometry, pH probe) for presenting symptoms (Figure 2c). The majority of patients had radiographic imaging performed, including CT of the neck (37.6%), chest x-ray (8.8%), and MRI of the neck (4%). Of note, 21.4% of the CT neck scans performed were reported as normal based on the radiology report. In addition, 33.6% of patients underwent pulmonary function testing; of these, 23.7% had a normal result.

Figure 2: Diagnostic evaluation of iSGS patients.

Physicians seen during first presentation for iSGS symptoms (a), studies performed during evaluation (b) and specialists seen prior to iSGS diagnosis (c) are depicted.

* Cardiac workup included echocardiograph, stress testing, and/or catheterization.

** GI workup included esophageal manometry, pH probe, and esophagogastroduodenoscopy

Table 2:

Physicians who made the diagnosis of iSGS

Physician	Number of Patients (%)
Otolaryngologist	98 (78.4%)
Pulmonologist	16 (12.8%)
Other Physician*	6 (5%)
Unknown	5 (4%)

^*Other physicians include radiologist, anesthesiologist, thoracic surgeon

Negative Consequences of Delayed Diagnosis

Patients were misdiagnosed with multiple respiratory diseases, including asthma, chronic sinusitis, or bronchitis/pneumonia. The majority of patients (54.8%) were misdiagnosed with asthma, 75% of whom were treated unsuccessfully with inhaled steroids prior to diagnosis. Interestingly, misdiagnoses were not associated with prolonged healthcare delay or total time to diagnosis. Seventeen patients had a total of 20 ED and urgent care visits. The number of ED visits was increased in patients with a healthcare delay greater than 6 months (p<0.05). There were 14 hospital admissions, 10 urgent dilations, and 9 procedures that might have been avoided with earlier diagnosis, including 4 tracheostomies (Table 3). Some ED visits and hospitalizations (25% and 50%, respectively) led to the diagnosis of iSGS (Figure 3). Additional procedures performed included Nissen fundoplication, septoplasty and laser lingual tonsillectomy, gastric inlet ablation, and cardiac catheterization, with each failing to resolve symptoms of dyspnea resulting in subsequent iSGS diagnosis.

Table 3:

Consequences of delayed iSGS diagnosis, including avoidable medications, additional procedures, emergency department/hospital admissions, and urgent dilation surgery.

Patient Consequence	% patients
Inhaled Steroid Treatment	49.2%
Antibiotic Treatment	15.2%
Additional Procedures  Tracheostomy       4   Cardiac Catheterization     2  Nissen Fundoplication     1  Gastric Inlet patch ablation  1  Septoplasty, Tonsillectomy   1	7.3%
≥1 Emergency Department Visit (20 total, 17 patients)	13.7%
≥1 Hospital Admissions (14 total, 13 patients)	10.5%
Urgent First Dilation Surgery	8.1%

Figure 3: Emergency Department and Hospitalization diagnoses.

Depiction of patient diagnoses assigned during emergency department visits (a) and at time of hospitalization (b).

Assessment of Risk Factors for Delayed Diagnosis

Patient factors such as sex, race, comorbidities, BMI, insurance type, and distance to treating medical center were not associated with increased time to first presentation, healthcare delay or total time to diagnosis. In analysis of patients diagnosed before and after the median diagnosis date (July 1, 2015) there were no significant differences between time to first presentation, healthcare delay, or total time to diagnosis (p=0.97, p=0.80, p=0.11).

DISCUSSION

Diagnostic delay in this iSGS patient cohort was significant, with a median total time to diagnosis greater than 2 years. This rare disease was commonly misattributed to asthma which resulted in a large number of individuals receiving inhaled steroids which did not relieve their dyspnea. Delays had consequences to patients, including emergency department visits, hospitalizations, urgent dilation procedures, and surgeries (including tracheostomy) that could have been avoided with earlier diagnosis. Earlier referral to otolaryngologists, CT neck imaging, and pulmonary function testing were factors that expedited iSGS diagnosis and treatment.

This study suggested a number of reasons for delay in diagnosis, including the rare incidence of iSGS, which is considered a rare disease by the National Organization for Rare Disorders and between 15-25% of all cases of subglottic stenosis.¹¹ In addition, iSGS often presents in otherwise healthy individuals with few comorbidities and likely few previous encounters with the health care system, which may explain the delay in first presentation. This is compounded by visits to primary care physicians, who are trained to provide care for a broad scope of conditions and may never have encountered iSGS in their practice or even be aware of the disease. In addition, the symptoms of iSGS are shared with more common pulmonary diseases including asthma, which has an incidence of 1 in 250 individuals, in stark contrast to iSGS with an estimated incidence of 1 in 400,000.^1,12 While iSGS is often encountered by otolaryngologists, it is a less considered cause of dyspnea than other conditions (i.e. asthma) more commonly encountered by pulmonologists and primary care physicians. While it is reasonable that many iSGS patients in this study were first treated for asthma, a prompt reassessment in non-responders may have led to an earlier correct diagnoses and relief of dyspnea.¹² In particular, pulmonary function tests and CT neck and/or chest were found to be helpful in identifying a fixed extrathoracic obstruction in patients in this cohort with 79% and 76% yield respectively.

Diagnostic delay has significant consequences for iSGS patients. In addition to prolonging patients’ dyspnea, delay results in unnecessary ED visits, hospitalizations, and surgeries, which result in unnecessary healthcare costs. Patients with a longer delay had a greater number of urgent encounters, as evidenced by the positive correlation between delay and patient ED visits. In addition, 9 individuals had procedures that would have been avoided with an earlier iSGS diagnosis, including Nissen fundoplication, cardiac catheterization, gastric inlet patch ablation, tonsillectomy, and septoplasty. These unnecessary surgeries highlight diseases that are confounded with iSGS during evaluation, including gastroesophageal reflux and chronic rhinosinusitis. Additionally, four patients in this cohort required tracheostomy which suggests these patients' airway stenosis was severe enough to require urgent airway intervention. These negative consequences significantly impact patient quality of life and may be prevented with an earlier assessment by physicians familiar with the disease.

An expected finding of this study is that early referral to an otolaryngologist was closely related to earlier iSGS diagnosis and treatment. Otolaryngologists were the primary specialists to diagnose and treat iSGS in this study with presentation to an otolaryngologist, iSGS diagnosis and first dilation procedure with a median time of 1.1 months. This is likely due to otolaryngologists’ familiarity with this rare disease and their routine use of flexible laryngoscopy/tracheoscopy, which was the most common study to assess airway narrowing. It should be recognized that referral to otolaryngology may be limited by referral patterns, lack of familiarity with the subspecialty, or lack of access to specialists in remote or resource-limited settings.¹³ In addition, 23% of this patient cohort experienced stridor as a presenting symptom, which is an indication for otolaryngologist evaluation of extrathoracic obstruction via laryngoscopy. Increasing awareness of iSGS during continued medical education of general practitioners may help physicians make appropriate referrals for stridulous and dyspneic patients with fixed extrathoracic obstruction who would benefit from further evaluation by an otolaryngologist.¹⁴

Diagnostic delay has been studied in the context of other otolaryngologic and respiratory diseases, highlighting causes and consequences of delayed referral. For head and neck cancer, reasons for diagnostic delay have been categorized, with the most frequent causes attributed to (1) time from referral to otolaryngology clinic appointment, (2) delay in referral of a symptomatic patient to otolaryngology clinic, (3) delay in employing an appropriate diagnostic test or procedure, and (4) delay in reporting a pathology/imaging result for an incidental lesion.¹⁵ This delayed diagnosis further impacts treatment initiation, which when prolonged beyond 60 days is associated with increased recurrence and worse survival.¹⁶ Patients with paradoxical vocal fold motion dysfunction experienced a mean time from symptom onset to diagnosis of 2.6 years (median 1 year), with consequences for the patient being inappropriately treated for asthma.¹⁷ For voice disorders of all types, delay in referral to an otolaryngologist is associated with increased healthcare costs.¹⁸ In respiratory diseases, 41% of idiopathic pulmonary fibrosis patients reported a misdiagnosis of respiratory symptoms, with a median delay of 2.1 years. Delay in referrals and treatment for these other diseases have shared themes with our iSGS patient population, including unnecessary medical and surgical therapy.

While this study demonstrated significant diagnostic delay in patients with iSGS, there are several limitations. Calculation of delay was limited to patients with sufficient data. In addition, initial symptom onset, as documented in the medical record, was subject to recall bias by patients which adds error to calculating time to first presentation and total time to diagnosis. However, healthcare delay, which was responsible for the greater portion of diagnostic delay, was more accurately calculated by objective data from electronic medical records. Further, patient histories and diagnostic studies often occur in parallel, making it challenging to retrospectively determine a singular event (i.e. CT scan, visit to emergency department) which led to iSGS diagnosis. Nevertheless, pulmonary function testing, CT imaging of the neck, and otolaryngology referral were all individually correlated to earlier diagnosis of iSGS. These steps represent opportunities to reduce diagnostic delay in patients with iSGS who face obstacles in diagnosis due to the rarity of the condition and challenges during evaluation.

Conclusion

The diagnosis of iSGS is frequently delayed, due to barriers in prompt patient presentation to healthcare provider and during evaluation. Individuals with iSGS are commonly misdiagnosed with asthma and do not respond to inhaled corticosteroids. Otolaryngologists were the primary specialists to diagnose iSGS and flexible laryngoscopy was the most common assessment technique. Earlier referral to an otolaryngologist and diagnostic testing in patients with continued respiratory symptoms (CT neck imaging, pulmonary function testing) were associated with reduced diagnostic delay in this patient population.