Wegener’s granulomatosis is a multisystemic disease characterized by foci of necrotizing vasculitis and granuloma formation. Subglottic stenosis may occur either as a presenting feature or a late-stage manifestation of the disease, but will occur in approximately 10-20 per cent of cases.
We present a series of seven cases of Wegener’s granulomatosis with subglottic stenosis and discuss our management of this condition. Where there is active disease, tracheostomy is the first- line surgical treatment of respiratory obstruction, as an adjunct to full medical therapy. More aggressive or elaborate surgical treatments should be reserved for non-active cases in which patients have not required medical treatment for one year.
Keywords: Wegener’s granulomatosis; Constriction, Pathologic; Trachea; Tracheostomy
Wegener’s granulomatosis is a necrotizing, granulomatous vasculitis that affects primarily the upper and lower respiratory tracts and kidneys.1 Besides these sites, there may be lesions in the oral cavity, larynx and trachea, ulceration of the skin, polyarthritis, and orbital involvement. Cranial nerve palsies may also occur. Laryngeal involvement is uncommon, but if present the subglottis and upper trachea are most commonly involved, with a risk of circumferential scarring and critical narrowing of the airway.
Subglottic stenosis occurs in approximately 10 to 20 per cent of patients with Wegener’s granulomatosis,2 who may present with this manifestation of the disease.3 It may be the only indication of disease, as was the case in half of the series of Langford et al.1 Symptoms range from cough and shortness of breath to life- threatening stridor. Initial diagnosis may be confused with the respiratory symptoms from co-existent pulmonary disease, and the stridor can easily be misdiagnosed as the wheeze of bronchial asthma. Where glottic involvement occurs, the voice may change.
The test for anti-neutrophil cytoplasm antibody (ANCA) is useful in diagnosis and is particularly helpful in unexplained subglottic stenosis. A positive ANCA test confirms the cause.4 C-anti- neutrophil cytoplasm antibody is cytoplasmic-staining and has a specificity of 99 per cent and a sensitivity of 91 per cent in active multisystem disease. P-anti-neutrophil cytoplasm antibody indicates perinuclear staining and is less commonly positive (between 20 and 40 per cent).5
We report a series of seven cases of Wegener’s granulomatosis with subglottic stenosis. The optimum treatment of such patients has not been formally ascertained, due mainly to the small number of sufferers.4 This uncontrolled, retrospective study looked at the airway management used, in an attempt to justify the rationale of our approach in light of published data concerning the natural history and alternative treatments for the condition. The study also looked at the correlation of the biopsy results with the ANCA test results during the course of the illness.
Materials and methods
The cases studied had been sent as secondary or tertiary referrals to the Queen Elizabeth Medical Centre, a regional teaching hospital in the United Kingdom. Wegener’s granulomatosis has long been a special clinical interest of the senior author, and cases were entered into a database (Paradox). Seven cases were treated over a 13-year period, all involving the subglottis. These were assessed retrospectively in an attempt to further characterize this unusual disease entity and to construct a rationale for diagnosis and management of the compromised airway. To this end, the existing literature on the subject was critically evaluated.
In order to identify relevant papers, a computer search (Medline) was used. Search criteria were Wegener’s granulomatosis (all fields) and subglottic stenosis (all fields).
Seven cases that fitted the inclusion criteria were referred to our department during a 13-year period. The diagnosis was made by histopathological examination of either nasal or renal biopsy tissue. Four patients had a positive C-ANCA test on presentation. Two of the patients had a negative ANCA test initially, but these became positive later in the disease. One patient had a positive P- ANCA test. The findings are summarized in the Table and in the case summaries.
No patient had isolated subglottic involvement. There were six female patients and one male patient. The age range of the group was between 20 and 62 years. Two patients had laryngeal symptoms at the time of presentation and five subsequently developed them between one and 12 years after the initial diagnosis of Wegener’s granulomatosis.
In two of the cases of delayed subglottic involvement the complication occurred despite the patient being on chemotherapy (case 4 was receiving cyclophosphamide and prednisolone and case 5 azathioprine and prednisolone). In the third case, the subglottis became involved after a long period of remission.
The diagnosis of subglottic stenosis was made on flexible pernasal laryngoscopy and confirmed by radiological imaging of the larynx. The severity of the subglottic stenosis was determined from clinical history, endoscopie examination, radiological measurement of the length of the stenosed segment and flow loop studies (the latter in two patients).
The severity of the patient’s symptoms, general health, activity of the Wegener’s granulomatosis and assessment of the severity of subglottic stenosis determined the further management plan.
All seven patients had active disease, and immunosuppressant and steroid treatment was needed almost continuously from the time of diagnosis. Four patients required a tracheostomy to relieve the airway obstruction. These patients had subglottic stenosis that was refractory to maximal drug therapy. Three patients who had less severe symptoms were treated with medical therapy alone and with regular clinic surveillance of their subglottic stenosis.
CLINICAL PRESENTATION AND AIRWAY MANAGEMENT OF THE PATIENTS
A female patient aged 32 years presented in 1989 with shortness of breath, nasal obstruction, tinnitus, deafness and arthralgia. A diagnosis of Wegener’s granulomatosis was established by nasal biopsy and a positive C-ANCA test. One year later, she developed voice changes and shortness of breath and was clinically diagnosed with subglottic stenosis. The subglottic stenosis was treated conservatively. The CANCA test remained positive throughout the illness. Initially, prednisolone and cyclophosphamide were used to treat the systemic manifestations of the disease. Her subsequent course was one of relapses and remissions treated by methotrexate and prednisolone. She received this combination therapy intermittently due to repeated relapses. She was treated by continuous methotrexate for five years. The patient continued to have repeated chest infections, arthralgia and blocked nose. Her C- ANCA test continued to be positive. The methotrexate was stopped two years ago and she was treated with prednisolone only because she was attempting to conceive. At the time of writing the patient was receiving prednisolone 5 mg on alternate days.
A female patient aged 34 years presented in 1993 with a hoarse voice and nasal obstruction with no other systemic symptoms. Nasoendoscopy showed subglottic stenosis and congested vocal cords. A PANCA test was positive for one year, following which it became negative. Nasal biopsy showed inflammatory granulation tissue without any granuloma formation. Magnetic resonance imaging (MRI) revealed a 3-cm segment of subglottic stenosis. Six months later the patient developed headache, weakness and paraesthesia of both the upper and lower limbs. In view of the severity of the breathing symptoms, tracheostomy was performed three years after the diagnosis. The patient’s medical treatment involved prednisolone and cyclophosphamide until she moved house and was transferred to another hospital’s care. During her follow-up period she experienced active ENT symptoms, with stridor, nasal blockage and recurrent epistaxis.
A female patient aged 53 years presented in 1998 with shortness of breath, hearing loss and nasal obstruction. An ENT examination revealed perforation of the right tympanic membrane and granulation tissue in the nose. The patient had ankylosing spondylitis. The C- ANCA test was positive and nasal biopsy showed typical necrotizing granulomata. Three years later she developed voice changes and a diagnosis of subglottic stenosis was established. An MRI scan revealed tracheal and subglottic stenosis. Tracheostomy was performed after attempts to treat the patient with laryngcal and trachal dilatation failed. The patient was also treated with prednisolone and cyclophosphamide and she continued to have repeated relapses. At the time of writing, her most recent treatment regime was weekly methotrexate injection and daily oral cyclosporine and prednisolone.
A female patient aged 66 years presented in 1990 with left-sided nasal obstruction and discharge. The clinical diagnosis was left- sided pansinusitis. Biopsy of the nasal granulation tissue showed typical necrotizing granulomatosis of Wegener. A C-ANCA test was negative at the time of presentation but became positive two years later. One year after presentation the patient developed respiratory and renal manifest\ations. At the age of 64 years the patient developed subglottic stenosis. A computerized tomography (CT) scan confirmed the diagnosis of mild subglottic stenosis. The patient received prednisolone and cyclophosphamide for four years and a tracheostomy was not required. The patient died in 1994 due to renal and pulmonary complications of the disease.
Computed tomography scans of the larynx and trachea of case 6. (a) Severe subglottic stenosis on the axial section; (b) the stenosis is shown to have an overall length of 3 cm. This patient was managed by tracheostomy.
A male patient aged 55 years presented in 1992 with weakness of the limbs, paraesthesia and renal impairment. He also had bilateral sensorineural hearing loss. One year later he developed voice changes, and the diagnosis of subglottic stenosis was established clinically. The diagnosis of Wegener’s granulomatosis was confirmed by renal biopsy. Tests for C-ANCA were negative for the first six years after presentation but later became positive. Tracheostomy was not performed. At the time of writing, the patient continued to receive prednisolone and azathioprine and had developed marked stridor.
A female patient aged 58 years presented in 1981 with acute renal failure due to glomerulonephritis. The C-ANCA test was positive and renal biopsy revealed Wegener’s granulomatosis. Twelve years later the patient presented with breathing difficulties, a hoarse voice and nasal obstruction. She was treated with prednisolone and azathioprine. Nine years after the diagnosis of Wegener’s granulomatosis, and with worsening breathing symptoms, laryngeal CT scan showed a severe 3-cm length of Wegener’s granulomatosis (see Figure). A tracheostomy under general anaesthesia was performed because of the severity of respiratory obstruction. Post- operatively, the patient continued to receive azathioprine and prednisolone.
A female patient aged 59 years presented in 1997 with nasal obstruction and rhinorrhoea, left-sided hearing loss, voice changes and mild stridor. Examination of the nose revealed a perforated nasal septum and fibre-optic endoscopy showed subglottic stenosis. The C-ANCA test was positive. Histopathological examination of the nasal septum and subglottic area showed non-specific inflammatory changes. The patient was diagnosed as having Wegener’s granulomatosis and treated with azathioprine and diclofenac. Her past history revealed chronic rheumatoid arthritis and coeliac disease, diagnosed 10 years before developing Wegener’s granulomatosis. Larynx MRI and CT scans confirmed the subglottic stenosis. Upon increasing stridor, a tracheostomy was performed three years after the diagnosis of Wegener’s granulomatosis. The patient developed severe voice changes and complete loss of voice two years later. Repeated dilatations of the subglottic area temporarily improved the voice for a few months. Laser excision was not attempted because of difficulty in performing tracheoscopy due to severe subglottic stenosis and also because of the patient’s cervical spine problems due to her rheumatoid arthritis. At the time of writing, the patient continued to receive azathioprine and diclofenac and was followed-up six-monthly.
Immunosuppression with drugs such as prednisolone, cyclophosphamide and azathioprine has revolutionized the treatment of Wegener’s granulomatosis, with a dramatic reduction in mortality. However, patients continue to suffer considerable morbidity from the disease itself and its treatment. The risk of relapse is high.3
Although over 90 per cent of patients with Wegener’s granulomatosis will have positive ENT findings at some stage in their disease,6,7 subglottic stenosis will occur in 10 to 20 per cent of all cases.2,7 Although two of the seven cases reported here presented with subglottic stenosis, none had disease confined purely to the subglottis. The sexes are equally afflicted by the generalized condition.3 However, we found that women greatly outnumbered men (6:1). A similar ratio has been noted previously.7,8 The reasons for this are unclear and suggest heterogeneity.
The differential diagnosis of subglottic stenosis includes carcinoma, tuberculosis, syphilis and other infectious diseases, lymphoma, and prolonged intubation.7,9 Tuberculosis may occur in susceptible individuals, such as patients receiving immunosuppression therapy, and we have encountered both diseases in one patient who had pulmonary tuberculosis in addition.
We found that C-ANCA was only positive in 4/7 cases at presentation; two cases became positive later. This was not constant throughout the course of the disease and was modified by medical therapy. The P-ANCA test was positive in one further case and mirrors the findings of a previous report.1 This suggests that both P-ANCA and C-ANCA should be routine tests and should be repeated when there is diagnostic uncertainty; they complement the biopsy.
A CT scan provides accurate assessment of tracheal lesions and was performed in five patients. The reformatted coronal image is particularly useful to evaluate vocal cord involvement.10 Stenoses measured up to 3 cm distal from the subglottis to extend substantially into the cervical trachea; this is similar to the experience of others.9
The diagnosis of isolated subglottic stenosis may be difficult histologically. Macroscopically, Wegener’s granulomatosis appears as a reddish, friable, circumferential narrowing just below the vocal cords. Histological section will typically show a more non-specific pattern of inflammation, as opposed to biopsies from other sites.9,11 Biopsy yield is low (5 per cent sensitivity).1 We found that histological confirmation was obtained most commonly from the nasal mucosa.
The subglottic lesion is atypical in a number of other ways. It is frequently resistant to systemic immunosuppressive therapy and fibrosis may develop during aggressive treatment. Systemic immunosuppression may not be indicated in isolated subglottic stenosis because of this.1 Although all our patients received medical treatment, there was no reliable long-lasting benefit to the course of the subglottic stenosis, and periods of remission were short-lived. This reinforces the opinion that subglottic stenosis can occur independently of the systemic disease and its treatment.
The goal is to eradicate airway obstruction whilst retaining natural airway anatomy. Mild or reversible lesions require no surgical intervention. Fixed lesions tend to require surgery, either tracheostomy, resection or dilatation. We found that tracheostomy was required in 4/7 cases, as was the case in another study (60 per cent).3 Decannulation is challenging because of continued disease activity and restenosis. There are limited reports of resection and re-anastomosis. In one report, successful healing was not followed by decannulation.12 In another report of three segmentai resection and re-anastomosis procedures, decannulation was only achieved in one case. Fears appear unfounded that immunosuppression interferes with wound healing, reactivates local disease and disturbs the anastomosis.8
Intralesional glucocorticoid has been used to treat subglottic stenosis and may reduce the need for systemic therapy. It is a safe and effective treatment when used in conjunction with intratracheal dilatation. It should ideally be used without concomitant systemic immunosuppression in local disease.1 However, multiple treatments are required, and this modality was not suitable for our patients. Results of laser excision are varied but generally do not allow decannulation of the patient.
We have found that subglottic Wegener’s granulomatosis is unpredictable and atypical because it targets women. The C-ANCA test was not always positive initially and should be repeated if the diagnosis is suspected. A combination of clinical, immunological and histopathological tests is needed to establish the diagnosis, and CT scanning helps to define its extent. The stenosis may extend 3 cm into the cervical trachea.
Surgical relief of the airway is not always necessary. A subset of patients has subglottic stenosis that is neither prevented by nor responsive to aggressive immunosupprcssant medication, and so tracheostomy should be performed promptly in these cases. Surgical resection by laser or reconstruction should be limited to the patients who are in remission and require no immunosuppressive therapy for at least a 12-month period. In our limited experience, laryngotracheal dilatation alone is not a reliable form of airway management. Combined therapy of intralesional corticosteroids and dilatation does offer the best hope for decannulation in isolated Wegener’s granulomatosis, but we have no experience of this.
* Subglottic stenosis and upper tracheal stenosis are well known complications of Wegener’s granulomatosis
* This uncontrolled, retrospective series reports the management of seven cases
* Conclusion: subglottic involvement may be difficult to diagnose and manage. Patients may not respond to immunosuppression; in such cases prompt surgical intervention may be required
1 Langford CA, Sneller MC, Hallahan CW, Hoffman GS, Kammerer WA, et al. Clinical features and therapeutic management of subglottic stenosis in patients with Wegener’s granulomatosis. Arthritis Rheum 1996; 39:1754-60
2 Waxman J, Bose WJ: Laryngeal manifestations of Wegener’s granulomatosis: case report and review of the literature. J Rheumatol 1986;13:408-11
3 Hoffman OS, Kerr GS, Leavitt RY, Hallahan CW, Lebovics RS, et al. Wegener’s granulomatosis: an analysis of 158 patients. Ann Intern Med 1992;116:488-98
4 Hoare TJ, Jayne D, Rhys Evans P, Croft CB, Howard DJ. Wegener’s granulomatosis, subglottic stenosis and antineutrophil cytoplasm antibodies. J Laryngol Otol 1989; 103:1187-91
5 Roland NJ, McRae RDR, McCombe AW. Key Topics in Otolaryngology,2nd edn. Oxford: Bios, 2001:17\0-1
6 Aozasa K, Ohsawa M, Tajima K, Sasaki R, Maeda H, Mastsunaga T, et al. Nation-wide study of lethal mid-line granuloma in Japan: frequencies of Wegener’s granulomatosis, polymorphic reticulosis, malignant lymphoma and other related conditions. Int J Cancer 1989; 44:63-6
7 Lebovics RS, Hoffman GS, Leavitl RY, Kerr GS, Travis WD, et al. The management of subglottic stenosis in patients with Wegener’s granulomatosis. Laryngoscope 1992:102:1341-5
8 McDonald TJ, Neel HB 3rd, DeRemee RA. Wegener’s granulomatosis of the subglottis and the upper portion of the trachea. Ann Otol Rhinol Laryngol 1982;91:588-92
9 Herridge MS, Pearson FG, Downey GP. Subglottic stenosis complicating Wegener’s granulomatosis: surgical repair as a viable treatment option. J Thorac Cardiovasc Surg 1996; 111:961-6
10 Screaton NJ, Sivasothy P, Flower CD, Lockwood CM. Tracheal involvement in Wegener’s granulomatosis: evaluation using spiral CT. Clin Radiol 1998;53:809-15
11 Yumoto E, Saeki K, Kadota Y. Subglottic stenois in Wegener’s granulomatosis limited to the head and neck region. Ear Nose Throat J 1997;76:571-4
12 Arauz JC, Fonseca R. Wegener’s granulomatosis appearing initially in the trachea. Ann Otol Rhinol Laryngol. 1982;91:593-6
A. ALAANI, F.R.C.S.ED., F.R.C.S.I., R. P. HOGG, M.PHIL., F.R.C.S. (O.R.L.-H.N.S.), A. B. DRAKE LEE, M.M.ED., PH.D., F.R.C.S.
From the Department of ENT, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH, UK.
Accepted for publication: 13 June 2004.
Address for correspondence:
Mr A.B. Drake Lee,
Consultant ENT Surgeon,
Department of ENT,
Queen Elizabeth Hospital,
Birmingham B15 2TH, UK.
E-mail: [email protected]
Mr A. B. Drake Lee takes responsibility for the integrity of the content of the paper.
Competing interests: None declared
Copyright Royal Society of Medicine Press Ltd. Oct 2004