METHODS
The study comprises a prospective cohort analysis comparing short course antimicrobial therapy +/- early surgical debridement versus prolonged course antimicrobial therapy alone in patients with NOE. The treatment protocol was approval by the Head & Neck Clinical Governance team at South Tyneside and Sunderland NHS Foundation Trust. All patients admitted with a clinical diagnosis of NOE to Sunderland Royal Hospital (SRH) between January-September 2019 were included.
Factors evaluated on admission included duration of symptoms, any relevant microbiology results, antibiotic therapy to date, history of ear syringing, patient pain score (1-10), cranial nerve palsies, and medical comorbidities (e.g. diabetes mellitus).Baseline investigations, including an ear swab, baselines blood tests (FBC, U&E, LFT, glucose, HbA1c, CRP, ESR), and a CT temporal bone, were performed for each patient.
In all cases the ear was examined by a senior member of the ENT team (registrar or consultant) and the canal was given a staging score (see Fig. 1):
Stage 1 = Normal or near normal ear canal.
Stage 2 = All of tympanic membrane visible, abnormal ear canal with granulation, polyp or swelling.
Stage 3 = more than 50% of tympanic membrane visible, due to polyp, granulation or swollen ear canal.
Stage 4 = less than 50% of tympanic membrane visible, due to polyp, granulation or swollen ear canal.
A significant improvement was defined as an improvement of 2 stages (i.e. stage 4 to stage 2, or stage 3 to stage 1).
A novel CT grading system was proposed to grade the degree of osteomyelitis (see Fig. 2):
Stage 1 = Soft tissue swelling , no osteomyelitis.
Stage 2 = Localised osteomyelitis to ear canal
Stage 3 = Evidence of inflammatory disease extending beyond ear canal, but not extending out of temporal bone.
Stage 4 = Disease extending beyond temporal bone OR presence of cranial nerve abnormalities.
If there were cranial nerve deficits, or the CT scan was stage 3/4, or the diagnosis was unclear, then an MRI scan was requested to evaluate soft tissue extent and meningeal involvement. Where the duration of antibiotics was to be extended beyond 6 weeks a gallium and bone scan was undertaken to extent of evaluate resolution and monitor treatment.
Patients were routinely commenced on intravenous piperacillin-tazobactam and gentamicin in combination. Review by a single surgeon (PA), based upon a database of previous NOE outcomes, determined whether patients were then enrolled on short course (2 weeks) +/- early debridement versus long course (6+ weeks) intravenous antimicrobial therapy.
Review of historical data (unpublished) suggested that the following parameters conferred a good prognosis:
Patients who met these criteria were enrolled on 2 weeks of intravenous (short course) antimicrobial therapy, and were subsequently switched to oral ciprofloxacin for a further 4-week course of treatment. All other patients were assigned to long course (6 weeks intravenous) antimicrobial therapy either had their entire treatment as an inpatient or administration in the community via a long line inserted prior to discharge.
Surgical intervention was considered only in selected circumstances; specifically:
A) Presence of an ear canal lesion that fails to respond to treatment after 1 week. In this case a biopsy with or without debridement wasoffered.
B) If at the end of treatment there was an area of necrotic bone that had not epithelised, or a bony sequestrum was present, then surgical removal wasoffered.
C) Early in course of disease for pain relief where this failed with improve with medical therapy alone.