2013/09/19 by Andy M. Courson, Holenarasipur R. Vikram, David M. Barrs · 16 citations
Medicine · #Ear Surgery and Otitis Media #Sinusitis and nasal conditions #Streptococcal Infections and Treatments
paper · pdf · doi:10.1002/lary.24093
What are the criteria for terminating treatment for necrotizing (malignant) otitis externa? The diagnosis of necrotizing, or “malignant,” otitis externa (NOE) is typically made by the clinical picture of a severe otitis externa, often in a diabetic or immunocompromised patient, with associated skull base soft tissue infection and osteomyelitis, as visualized on magnetic resonance (MR) or computed tomographic (CT) scanning. While treatment efficacy has improved since the condition was first described decades ago, NOE remains a severe infection with potential for significant morbidity and mortality. Bacterial NOE is ideally treated with at least 6 weeks of culture-directed systemic antibiotics, often with infectious disease consultation. Patients presenting with NOE frequently have been previously treated with oral or topical antibiotics, rendering cultures negative. In these cases, many clinicians empirically treat with antipseudomonal antibiotics. Longer therapy may be required with multidrug resistant strains of bacteria such as Pseudomonas aeruginosa, or with molds such as Aspergillus fumigatus. As NOE can mimic squamous cell carcinoma of the external ear canal on physical exam and CT scan, some clinicians biopsy granulation tissue when diagnosis is uncertain or if granulation tissue is present after 6 weeks of treatment.1 Once clinical exam and laboratory markers have normalized, the otolaryngologist faces the problem of how to determine if treatment has been adequate and can be discontinued with minimal risk of recurrence. As soft tissue involvement is expected to have resolved, the concern is for persistent osteomyelitis. Two questions are important: what is the appropriate duration of antibiotics and how can the clinician tell when the infection is resolved. The recommended treatment for osteomyelitis of peripheral bones is surgical debridement and parenteral or oral antibiotic therapy for 4 to 6 weeks.2 The optimal duration of treatment is unknown, but the rationale is based on a period of 3 to 4 weeks for bone to revascularize. In NOE, surgical debridement is not recommended for initial treatment as the central skull base is largely inaccessible. Antimicrobial therapy is ideally based on cultures, usually from the ear canal as culture from the tympanic bone or skull base is often impractical. A recent Cochrane Collaboration2 comparing parenteral and oral antibiotic treatment for chronic osteomyelitis noted that the length of treatment for chronic osteomyelitis has not been defined. With respect to oral versus parenteral antibiotics, the authors concluded that the method of antibiotics administration does not affect the rate of disease remission as long as the bacteria are sensitive to the antibiotic used. Additionally, in a recent invited review in Clinical Infectious Diseases, Spellberg and Lipsky3 examined articles in the English literature from 1970 to 2011 containing the term, “osteomyelitis.” These authors stated that no strong evidence supported treatment longer than the standard 4 to 6 weeks, that there were no good markers for successful treatment, and that relapse rates remain substantial. The answer to the first question regarding the optimal length of therapy for osteomyelitis in NOE, as with other areas of osteomyelitis, remains unanswered. For this reason, radiologic studies are largely employed to guide termination of therapy. When imaging for resolution of NOE, it is important to distinguish chronic bone and soft tissue changes from active infection. MRI and CT are frequently used in the diagnosis of osteomyelitis. MRI has superior detection of bone marrow edema, which precedes the cortical bone erosion best seen on CT. However, these changes can persist long after therapy, limiting their usefulness in determining the adequacy of antibiotic treatment. Al-Noury and Lotfy4 reviewed 18 patients with NOE, with CT and MRI performed at diagnosis and 6 and 12 months after initiation of treatment. On CT, 100% of patients had persistent cortical bone changes. On MRI, 60% of patients had soft tissue changes and 33% had marrow abnormalities that persisted at 1 year. Nuclear medicine studies using radiopharmaceuticals can help define the extent and eradication of infection. Three-phase bone scanning with Technetium-99m-hydroxymethylene-diphosphonate (99mTc-HDP) or Technetium-99m-methylene diphosphonate (99mTc-MDP) has traditionally been used for the initial diagnosis of osteomyelitis by identifying sites of increased osteoblast activity. Although sensitive for osteomyelitis, these scans are unable to differentiate between active infection and bone remodeling. Therefore, Gallium-67-citrate (67Ga) has been frequently employed to monitor treatment response. 67Ga scans show areas of infection, but they return to normal with disease resolution.4 An alternative to 67Ga scanning is 111Indium (111In), labeled leukocyte scintigraphy, which identifies neutrophil-mediated inflammation. Single-proton emission computed tomography (SPECT) is a nuclear medicine study that creates three-dimensional pictures and uses radiopharmaceuticals such as 111In or 67Ga. Combining CT with SPECT (SPECT/CT) allows for better anatomic localization. Djalilian and colleagues1 conducted a retrospective case review of eight patients with NOE. Initial diagnosis was made with a 99mTc bone scan and a 67Ga scan. Patients had a follow-up 67Ga scan after 6 weeks of antibiotics. All patients had normalized scans and antibiotics were discontinued. There were no recurrences, with a median follow-up period of 6 months. A recent meta-analysis5 of 23 studies looked at accuracy of plain films, MRI, CT, 99mTc bone scintigraphy, leukocyte scintigraphy, 67Ga scintigraphy, combined 99mTc bone and leukocyte scintigraphy, combined bone and 67Ga scintigraphy, and fludeoxyglucose positron emission tomography (FDG PET). FDG PET was found to have the highest accuracy for confirming or excluding chronic osteomyelitis. However, the specificity was not significantly higher than combined 99mTc bone and leukocyte scintigraphy or 67Ga scintigraphy. No studies of significance have looked at FDG PET specifically for the diagnosis and follow-up of NOE. Although there is some question as to whether FDG PET maintains its accuracy in the assessment of osteomyelitis in diabetics secondary to impaired glucose uptake, this may be found to be a viable imaging option for NOE in the future. Based on recommendations for the treatment of osteomyelitis, bacterial NOE should be treated with at least 6 weeks of antibiotics. Even with clinical resolution, there may be concern for residual infection in a surgically inaccessible skull base area. Imaging studies can therefore be considered. MRI, CT, and Tc99 cannot be used to document resolution as bone changes persist after disease resolution. Ga67 scintigraphy or 111In-labeled leukocyte scintigraphy combined with SPECT/CT can help differentiate healing bone from active infection. If these are positive, antimicrobial therapy is continued, and the scans can be repeated at 2 to 4 week intervals until normalization. No current evidence exists for the efficacy of FDG PET in imaging NOE, but the otolaryngologist should consider consultation with radiologists specialized in nuclear medicine as imaging modalities may change rapidly. Recurrence has been reported to be around 15% to 20%,6 so at least a 6 month follow-up for NOE with a low threshold for reimaging is ideal. Future research directed at determining the sensitivity and specificity of the various radiologic modalities in identifying resolution of disease could help unify recommendations for when to terminate therapy for NOE. NOE is a rare clinical entity and no large studies exist. Much of the literature focuses on accurate radiologic diagnosis. The studies presented were level 4 and level 3 evidence.