A surgical protocol for sinogenic brain abscess: the Oxford experience and a review of the literature

BACKGROUND

Rhinosinusitis-induced brain abscesses are rare but can result in devastating long-term sequalae and mortality; they require a high index of suspicion with early imaging to start early empiric parenteral antibiotic treatment covering aerobes and anaerobes.

METHODOLOGY

Our study was a retrospective analysis on 32 patients who were treated at Oxford University Hospitals for rhinosinusitis-induced brain abscess between February 2013 and June 2020.

RESULTS

Mean age of presentation was 45.83 for adults and 11.14 for children. Subdural collection was the most frequent abscess but 25% of patients had multiple sites of collection; the majority were in the frontal lobe. The most commonly identified pathogens were Streptococcus milleri group and Staphylococcus aureus; 93.75% of the patients were treated with combined Ceftriaxone and Metronidazole for an average of 8 weeks.

CONCLUSIONS

In our series most patients received also a prompt and aggressive surgical treatment with combined neurosurgical and ENT procedures in the majority; this was especially important in case of subdural empyema, Streptococcus milleri infection and direct intracranial spread of infection. More than half of the patients were treated with a single surgical procedure. Despite aggressive treatment, one third of patients experienced long-term neurological sequelae; there were no deaths.

P14.85 Impact of the neuro-radiologist and neuro-surgeon in contouring with the neuro-oncologist on local relapse rates for brain metastases treated with stereotactic radiosurgery

BACKGROUND

The audit evaluates the value of MDT, including neuro-radiologist and neuro-surgeon, review of contouring carried out by a clinical oncologist in stereotactic radiosurgery (SRS).

MATERIAL AND METHODS

A sequential audit was conducted of all patients receiving intracranial SRS at our local institution for the first 22 months of a new SRS service. Lesions were contoured first by clinical oncologist then reviewed/edited by the MDT. The initial contour was compared with final contour using Jaccard conformity and geographical miss indices. The dosimetric impact of a contouring change was assessed using plan metrics to both original and final contour. The impact of the contouring review on local relapse, overall survival and radio necrosis rate was evaluated with at least 24 months follow up (24–46 months).

RESULTS

113 patients and 142 lesions treated over 22 months were identified. Mean JCI was 0.92 (0.32–1.00) and 38% needed significant editing (JCI<0.95). Mean GMI was 0.03 (0.0–0.65) and 17% showed significant miss (GMI>0.05). Resection cavities showed more changes, with lower JCI and higher GMI (p<0.05). There was no significant improvement on JCI or GMI shown over time. Dosimetric analysis indicated a strong association of conformity metrics with PTV dose metrics; a 0.1 change in GTV conformity metric association with 6–17% change in dose to 95% of resulting PTV. Greater association was seen in resection cavity suggesting the geographical nature of a typical contouring error gives rise to greater potential change in dose. Clinical outcomes compared well with published series. Median survival was 20 months and local relapse free rate in the treated areas of 0.89 (0.8–0.94) at 40 months, and 0.9 (0.83–0.95) radio-necrosis free rate at 40 months with a median 17 months to developing radio-necrosis for those that did.

CONCLUSION

This work highlights that a MDT contour review adds significant value to SRS and the approach translates into reduced local recurrence rates at our local institution compared with previously published data. Radio-necrosis rates are below 10%. No improvement in clinical oncologist contouring over time was shown indicating a collaborative approach is needed regardless of experience of clinical oncologist. MDT input is recommended in particular in contouring of resection cavities.