Introduction
Necrotising fasciitis is a well recognised, life-threatening bacterial infection. Its aetiology is usually either polymicrobial, with mixed aerobic and anaerobic bacteria, or monomicrobial due to group A streptococcus.
Epiglottitis is an increasingly uncommon condition which can occur at any age and is often caused by Haemophilus influenzae serotype b (Hib). This organism is known to cause many infections, including cellulitis, but has only exceptionally rarely been implicated in necrotising fasciitis, and never in association with epiglottitis.
We report a case of acute epiglottitis in an adult, complicated by monomicrobial necrotising fasciitis; the causative organism in both instances was Hib.
Case report
A 64-year-old woman was referred to the ENT surgeons with a 24-hour history of worsening sore throat and difficulty swallowing. She had noticed a raised temperature, sweating and nausea, and had taken little orally over the previous 24 hours. She had a past medical history of hypertension, for which she took atenolol and amlodipine. Her other medications were naproxen (for osteoarthritis), omeprazole and furosemide. She was a non-smoker and lived independently with her husband.
On initial examination, the patient appeared unwell but had no hoarseness or stridor, and her vital signs were normal. Her neck was tender.
Blood tests showed raised concentrations of C-reactive protein (195 mg/l) and urea (10.5 mmol/l), with normal creatinine (109 µmol/l). The patient's electrolytes, full blood count and liver function tests were normal.
Flexible nasendoscopy performed on the ward demonstrated a red, inflamed epiglottis. Blood cultures were sent, and the patient was treated for presumed infective epiglottitis, with intravenous ceftriaxone, nebulised saline, oral co-codamol and intravenous fluids.
Blood cultures grew Gram-negative bacilli within hours, at which point the antibiotics were changed to ceftriaxone, metronidazole and clindamycin.
Within 18 hours, the patient developed septic shock refractory to fluid resuscitation, and was admitted to the intensive care unit. She was oliguric, and repeated blood tests showed deterioration in her renal function (urea 14.2 mmol/l and creatinine 134 µmol/l). She had a mild coagulopathy (prothrombin time 17 seconds, activated partial thromboplastin time 36 seconds, fibrinogen 492 mg/dl and platelet count 128 × 109/l) and a metabolic acidosis (H+ 51.2 nmol/l, pCO2 5.7 kPa and base excess −6.1 mmol/l). The serum lactate concentration was slightly elevated (2.2 mmol/l) and the central venous oxygen saturation was low (65 per cent), in keeping with hypoperfusion secondary to septic shock. Despite the general deterioration in the patient's condition, her airway remained patent, with no stridor or other evidence of obstruction.
The patient received supportive treatment, with further intravenous fluids, infusions of adrenaline and noradrenaline, and intravenous hydrocortisone, as per the local policy for septic shock.
An area of erythema approximately 10 × 15 cm had previously been noted over the patient's sternal area. This was initially tender and swollen, with a central area which appeared ‘bruised’. The patient had first noticed it when changing into her hospital gown. On admission to the intensive care unit, this area had increased in size and now (18 hours after hospital admission) had three dusky areas but no crepitus or blistering. Following review by the surgical team, immediate surgery was planned in view of probable necrotising fasciitis.
An uneventful inhalational induction was performed in the operating theatre, in the presence of the ENT consultant. Although tracheal intubation was straightforward, the epiglottis was noted to be inflamed, with localised pus and marked surrounding oedema. A swab of the area was sent to microbiology. The affected skin and fat of the sternal area was thoroughly debrided. This excised tissue was noted to have early signs of necrotic change, and was oozing fluid. There was no apparent continuity between the necrotic area and the inflamed epiglottis.
Post-operatively, the patient was transferred back to the intensive care unit, where she remained intubated and ventilated for approximately 48 hours. During this period, she made a steady improvement, with resolution of the acidosis and renal impairment and cessation of haemodynamic support. The wound was reviewed but further debridement was not required. Forty-eight hours after intubation, direct laryngoscopy showed reduction of the epiglottic swelling, there was a leak on deflation of the tracheal tube cuff, and the patient was successfully extubated.
Blood cultures taken on admission were found to contain H influenzae type b (Hib). This organism alone was also isolated from both the epiglottic swab and the debrided tissue. On the advice of the microbiologists, the patient was given a four-day course of oral rifampicin, and Hib vaccination was planned to clear any carriage of haemophilus.
Three weeks after her admission, the patient was transferred to the regional plastic surgery unit, where her wound was successfully grafted.
At her ENT clinic review six weeks later, the patient was found to be very well, with her graft healing satisfactorily and no abnormality on examination of her throat.
Discussion
Once commonly encountered in ENT and anaesthetic practice, epiglottitis has become a rare condition which many trainees will never have seen. Consultants may not have been involved with a case for some time. Until recently, the commonest cause was infection with H influenzae type b (Hib), but the condition can be caused by other bacteria, viruses and non-infectious agents, such as heat.Reference Gorelick and Baker1 Classically a disease of early childhood, routine immunisation against Hib in UK children has dramatically reduced the incidence of epiglottitis in this age group, leading to a relatively increased incidence in older children and adults.2, Reference Shah, Roberson and Jones3 An awareness of the possibility of the diagnosis in this population is therefore important.
Our case illustrates the typical features of epiglottitis: a brief history of severe sore throat, dysphagia and fever, and rapid development of marked systemic illness. Although less likely, due to their larger-diameter airway, adults may nonetheless develop complete airway obstruction and thus require close observation, with early intubation if the airway is thought to be at risk.
Our patient's ‘textbook’ presentation with epiglottitis was complicated by the highly unusual development of associated necrotising fasciitis. Necrotising fasciitis is part of a spectrum of necrotising soft-tissue infections. The term describes infection and necrosis of fascia and subcutaneous tissue, with relative sparing of underlying muscle and overlying skin. It can progress very rapidly and leads to extensive tissue destruction, systemic toxicity and death if not recognised and treated early with surgical debridement. Necrotising fasciitis is classed as type I or type II according to clinical and microbiological factors, as follows.Reference Stevens and Rose4
• Epiglottitis is increasingly uncommon in the UK due to routine childhood vaccination against Haemophilus influenzae type b (Hib)
• Necrotising fasciitis is a potentially fatal infection which, when monomicrobial in origin, is usually caused by group A streptococcus
• Epiglottitis has occasionally led to localised necrotising fasciitis, but this has never been reported in cases caused by Hib
• Necrotising fasciitis is now known to be a potential complication of Hib epiglottitis
• This may arise by haematogenous rather than local spread
• The presented case supports a single other report of Hib as a cause of monomicrobial necrotising fasciitis
Type I refers to a polymicrobial infection which tends to occur in patients with underlying risk factors, such as diabetes, peripheral vascular disease, advanced age or intravenous drug use. There is usually a clear portal of entry for bacteria, and subcutaneous gas is a common finding.
Type II refers to a monomicrobial infection, almost always with group A streptococcus, although, more recently, community-acquired methicillin-resistant Staphylococcus aureus has been reported as a cause.Reference Miller, Perdreau-Remington, Rieg, Mehdi, Perlroth and Bayer5 It more often occurs in previously healthy individuals of any age. There may be no obvious portal of entry for bacteria, in which case infection may arise by haematogenous spread of organisms from another site to damaged tissue. Subcutaneous gas is less common.
Our patient's case best fits the features of type II disease. She was previously well and had a monomicrobial infection. On surgical exploration, there was no apparent communication between the epiglottic and the necrotic areas, leading us to believe that the soft tissue infection had arisen by haematogenous spread.
An association between Hib epiglottitis and necrotising fasciitis has not previously been described. The Medline, Embase and Pubmed databases were searched without date restriction for English language articles containing the terms ‘Haemophilus influenzae’ or ‘epiglottitis’ combined with either ‘necrotising fasciitis’ or ‘necrotizing fasciitis’ within the title, abstract or keywords. Articles retrieved were reviewed for relevance. Only one other case of necrotising fasciitis caused by Hib was found, associated with leg cellulitis in a 10-month-old child, who recovered following surgery and antibiotics.Reference Collette, Southerland and Corrall6 A different H influenzae serotype (serotype f) has been reported as the cause of fatal necrotising fasciitis in the lower limb of a 65-year-old man.Reference McLellan, Suvarna and Townsend7
There are a handful of reports of epiglottitis leading to necrotising fasciitis by local spread to the neck, chest or mediastinum. Two of these cases were polymicrobial infections in adults, both of whom made a full recovery.Reference Kraus, Gatot, Leiberman, Nash and Fliss8, Reference Caballero, Sabater, Traserra, Alos and Bernal-Sprekelsen9 Three others were monomicrobial infections: an adult in whom peptostreptococcus species were isolated, another adult with Aeromonas hydrophilia, and a child with group A streptococcus.Reference Nguyen and Leclerc10–Reference Slack, Allen, Morrison, Garren and Roback12
The current case represents the first report of Hib epiglottitis leading to necrotising fasciitis, and adds to the small but important body of evidence regarding unusual causes of necrotising fasciitis. This case emphasises the need to be vigilant for evolving complications, however unexpected, in critically ill patients, and also the importance of good communication between all relevant specialties. Perhaps most importantly, however, this case highlights the features of two uncommon but life-threatening disorders, in which a good outcome is crucially dependent upon recognition and timely, appropriate management.
It is hoped that this report will benefit patients by raising clinicians’ awareness of these conditions, their initial management and possible aetiology.
Acknowledgements
I would like to thank Dr Jane Chestnut and Mr Peter Wardrop for their advice regarding the preparation of this report. Consent for publication was granted by the patient.
