Supraglottic laryngeal sarcoidosis masquerading as supraglottitis

Abstract

Sarcoidosis is a rare multisystem disorder of unknown aetiology characterised by non-caseating granulomas in involved organs; it is a diagnosis of exclusion. Laryngeal involvement affects only 0.5%–5% of those with sarcoidosis. It is an uncommon but important cause of supraglottic inflammation and oedema and should be considered in the differential diagnosis in patients with supraglottitis. This case describes a 30-year-old man who presented with stridor and shortness of breath. Flexible nasendoscopic examination revealed a grossly oedematous, pale pink, diffusely hypertrophied epiglottis. Surgical biopsy revealed non-caseating granulomatous inflammation. In the context of exclusion of hepatitis, anti-neutrophil cytoplasmic antibody (ANCA) positivity, malignancy and mycobacterial infection, the diagnosis of supraglottic laryngeal sarcoid was made. He is being treated with azathioprine immunosuppression with symptomatic improvement.

Background

Supraglottic inflammation and oedema is most often secondary to infectious supraglottitis. However, in a small number of cases, granulomatous pathology is the underlying cause. Laryngeal sarcoidosis is rare, affecting only 0.5%–5% of those with systemic sarcoidosis. Despite it having characteristic histological appearances of non-caseating granulomas, it remains a diagnosis of exclusion.

We present a case of supraglottic laryngeal sarcoidosis, initially diagnosed as infective supraglottitis. Only years later during a recurrent exacerbation of the disease was the diagnosis of laryngeal sarcoidosis made. This case serves as an important reminder to consider granulomatous pathology within the differential diagnosis of those with supraglottitis to ensure appropriate investigation and management of these patients.

Case presentation

A 30-year-old man presented to the emergency department (ED) with a 4-week history of worsening stridor, cough, odynophagia and inability to clear secretions. In the week prior to attendings ED he also mentioned onset of subjective fevers, chills, hoarse voice and progressively worsening shortness of breath on exertion.

Some weeks before, he had attended his General Practitioner with worsening odynophagia and respiratory distress and was referred to a private Ear, Nose and Throat (ENT) Surgeon. He was placed on the waitlist for a laser supraglottic resection due to concern regarding airway patency and for histopathological diagnosis. He was known to this ENT surgeon in the context of an episode of supraglottitis 6 years prior. At that time, he experienced 12 months of worsening stridor and dyspnoea. After failing initial medical therapy with intravenous antibiotics, he underwent a laser debulking of the lingual epiglottis and arytenoids in 2013 with histopathology revealing non-specific granulomatous inflammation and microscopy and culture growing Actinomyces and Group B Streptococcus. He tested negative for tuberculosis, fungal organisms and had a negative autoimmune and vasculitic screen. His CT neck showed an oedematous supraglottic larynx and CT chest was unremarkable. He was diagnosed with supraglottitis secondary to bacterial infection and given a 1 month’s course of oral antibiotics for Actinomyces eradication. He remained stable in the community with some improvement but never fully recovered. For the following 6 years until presentation to ED, he did not require hospital attendance.

His medical history was remarkable for type 1 diabetes mellitus, on a basal bolus insulin regime. He had allergies to morphine, cefaclor and codeine. He worked as a carpenter, lived at home with his partner, drank alcohol rarely and was a non-smoker.

Examination by the ENT surgical team in ED revealed a soft stridor and a hoarse voice. He was afebrile, haemodynamically stable, had a clear chest to auscultation and was saturating normally on room air. Flexible nasendoscopy (FNE) revealed an inflamed and grossly oedematous epiglottis with the true vocal cords only partially visible.

Blood tests in ED showed a raised white cell count of 21×10⁹/L, C-reactive protein (CRP) 45 mg/L and he had a normal chest X-ray.

He was admitted to the ENT surgical ward for observation and commenced on three times a day intravenous dexamethasone 8 mg to good effect, with improvement in stridor and work of breathing. Despite the progress in airway management, this proved problematic for his diabetic control. He was consented and booked for an urgent panendoscopy and lingual epiglottis laser resection to both gain a tissue diagnosis for the cause of supraglottic inflammation and reduce the bulk of the epiglottis improving airway patency. These biopsies were sent for histopathology and microbiology investigation. Following the procedure, he recovered well with an uncomplicated postoperative course. He was discharged home with a tapering course of oral prednisolone and plan for clinic follow-up.

His surgical histopathology demonstrated chronic non-caseating inflammation, granuloma and fibrosis with normal oral flora on microscopy (Group B Streptococcus and filamentous bacteria). These bacteria were not considered pathological as they were identified only in the superficial epithelial layers of the histopathology sample, suggesting colonisation. Cultures were negative for acid fast bacilli, fungi and PCR was negative for Chlamydia trachomatis and Neisseria gonorrhoea e.

Two days following discharge from hospital he experienced worsening of symptoms; dysphagia to solids, odynophagia, shortness of breath, drooling and trismus. He attended the ENT clinic earlier than planned. He had a stable airway but in the context of recent surgical intervention was commenced on a course of oral antibiotics (Augmentin Duo Forte) and booked for clinic review the following week.

On clinic review, his symptoms had again worsened and he described some central chest tightness. Examination revealed a soft stridor, dysphonia, trismus and FNE a grossly oedematous supraglottis with only partial view of the vocal cords (figure 1). His airway was stable however and saturating normally with no oxygen support. He was admitted and commenced on intravenous antibiotics (benzylpenicillin and metronidazole), dexamethasone 8 mg and a proton pump inhibitor.

Figure 1.

Flexible nasendoscopy showing a grossly oedematous pale pink epiglottis and limited view of the glottic larynx.

ECG on admission was normal however a raised D-dimer blood test in the context of chest tightness led to a CT chest being done, which was clear of pulmonary embolus but revealed considerable bilateral mediastinal and paratracheal lymphadenopathy (figures 2 and 3). CRP was raised at 72 mg/L though white cell count was normal.

Figure 2.

CT chest (coronal, post contrast, pulmonary arterial phase) showing bulky subcarinal and paratracheal lymphadenopathy.

Figure 3.

CT chest (axial, post contrast, pulmonary arterial phase) showing bulky right mediastinal lymphadenopathy.

He was reviewed by the infectious diseases and Immunology teams. Further investigations were negative for Epstein Barr virus, Cytomegalovirus, hepatitis, HIV, autoimmune/vasculitis screen (negative antinuclear antibody, ANCA, C3, C4, Ig levels, serum free light chains and ACE) and tuberculosis. In the context of the macroscopic and histological appearance of the supraglottic tissue, initial positive response to steroid therapy, mediastinal lymphadenopathy and relapsing course, both specialties felt the most likely diagnosis to be supraglottic laryngeal sarcoidosis, with potential mediastinal involvement. The patient improved with steroid therapy on the ward and was discharged home following inpatient respiratory review.

Outcome and follow-up

His symptoms continued to improve in the outpatient setting; on early follow-up he had no stridor, no laboured breathing and improved vocal quality.

Outpatient positron emission tomography (PET) scan demonstrated epiglottic uptake consistent with known inflammatory supraglottitis but there was no uptake in the neck, thorax or abdomen to suggest ongoing active sarcoid disease elsewhere.

The patient underwent bronchoscopic biopsy some weeks later. The histopathology of the biopsied level 7 carinal lymph node was of non-caseating granulomatous inflammation with appearances in keeping with sarcoidosis. This sample was again negative for acid fast bacilli and fungi but identified actinomyces species bacteria, though this was again considered to represent a superficial colonising organism by the infectious disease physicians and the microbiologists involved in this case. Flow cytometry of the sample was negative for lymphoproliferative disorders.

He was seen for follow-up in outpatients by Immunology. Given multiple biopsies of non-caseating granulomatous inflammatory tissue and the exclusion of hepatitis, ANCA positivity, malignancy and mycobacterial infection, they were in agreeance with a diagnosis of supraglottic laryngeal sarcoidosis. With long term steroid therapy contraindicated secondary to type 1 diabetes, the patient was commenced on a trial of azathioprine immunosuppression, commencing at 50 mg daily and increased to 150 mg daily in the coming months. He has responded symptomatically very well to therapy and has ongoing follow-up with immunology, ENT surgery and respiratory medicine for management of his supraglottic sarcoid.

This case describes a rare but important cause of supraglottitis. When assessing a patient with stridor and supraglottitis it is important to consider not only infectious, but also autoimmune, vasculitic and granulomatous aetiologies.

Discussion

Supraglottitis or epiglottitis is an inflammatory condition of the epiglottis and adjacent supraglottic structures.¹ It can cause significant supraglottic swelling and oedema and without prompt treatment, can potentially progress to life threatening airway obstruction. This condition is most commonly infectious in aetiology however other causes include trauma (thermal, caustic, foreign body), graft versus host disease following organ or bone marrow transplant, and granulomatous conditions such as vasculitis or exceedingly rarely, sarcoidosis.²

Sarcoidosis is a multisystem granulomatous disorder of unknown aetiology.³ It is characterised by non-caseating granulomas in involved organs and tissues and remains a diagnosis of exclusion.^{3 4} It typically affects young adults and has an estimated prevalence of 10–35 cases per 100 000 population.³ The disease is usually systemic, most commonly affecting the respiratory tract, however in 30%–60% of cases, patients are asymptomatic.³ Laryngeal involvement of sarcoid is rare, affecting only 0.5%–5% of those suffering with the disease.^{4 5} It typically presents with supraglottic involvement and appears as a pale pink, oedematous, diffusely hypertrophied epiglottis and associated supraglottic structures.^{4 6–8} This can be very easily mistaken for infectious supraglottitis or many other causes for supraglottic inflammation. Diagnosis is made by characteristic appearance of the larynx on endoscopic examination, histology of laryngeal biopsies showing non-caseating granulomatous inflammation and the exclusion of other granulomatous diseases (for example granulomatosis with polyangiitis or tuberculosis).^{4 7}

The mainstay of therapy is usually long-term oral steroids however some research suggests intralesional injection of steroids may have positive outcomes for laryngeal sarcoid involvement.^{4 7 9} Immunosuppressant therapy is employed where failure of steroid therapy occurs or where there is a contraindication to steroids.⁴ Surgical options include CO₂ laser debulking of the lesion or in extreme cases of airway compromise a supraglottic laryngectomy.⁴

A search of the PubMed database using the key term ‘laryngeal’ OR ‘supraglottic’ AND ‘sarcoid’ revealed eight case reports and two case series, describing 23 patients with laryngeal sarcoidosis.4 6 8–15 Almost all cases described supraglottic involvement. Many of these studies were published in the 1980s–1990s, one of which followed a similar course to the case described here, though treated with a supraglottic laryngectomy rather than conservative therapy.¹⁴ There were no articles noted on the Cochrane database specifically addressing laryngeal sarcoid. One case report described the use of azathioprine in the context of failure of steroid therapy.⁴ To our knowledge, this is the second ever reported case of the use of azathioprine in the treatment of sarcoid of the larynx.

This report describes a case of long standing insidious laryngeal sarcoidosis. His index exacerbation was treated as an infectious supraglottitis. He then remained without specific sarcoid therapy for many years before having a further exacerbation which led to the eventual diagnosis and appropriate management. This case illustrates the importance of considering all causes of supraglottic inflammation to ensure patients are investigated and managed appropriately.

Learning points.

• Sarcoidosis is a rare multisystem disorder of unknown aetiology characterised by non-caseating granulomas in involved organs; it is a diagnosis of exclusion. First line treatment is with oral steroids.

• Laryngeal involvement affects only 0.5%–5% of patients with sarcoidosis.

• This case describes a protracted disease course with multiple presentations over 6 years prior to a diagnosis of laryngeal sarcoid being made.

• It serves as a reminder that non-infectious causes of supraglottic inflammation and swelling exist. Further, it is important to consider granulomatous conditions in the differential diagnosis for patients presenting with this endoscopic finding.