Descending necrotising mediastinitis can complicate oropharyngeal infection and has a high associated mortality. We present three cases treated in our department and propose a treatment algorithm based on our experience and literature review. The primary oropharyngeal infection was peritonsillar abscess in two cases and odontogenic abscess in one. Two patients underwent cervicotomy and later thoracotomy. The third underwent cervicotomy with transcervical mediastinal drainage and later required pericardial drainage via a subxiphoid incision. All recovered fully and were discharged within 6 weeks. To enable successful treatment, diagnosis needs to be prompt and surgical drainage adequate. Thoracic management of the chest is essential.

Pharyngo-laryngo-oesophagectomy and gastric pull-up (PLOGP) is a complex and relatively uncommon procedure. The aim of this study is to analyse the results of PLOGP in patients with post-cricoid and cervical oesophageal squamous cell carcinomas. This study was a retrospective review of 26 patients (11 males + 15 females, mean age 63.5 years) who underwent PLOGP from 1988 to 1997. Eighteen (69 per cent) patients were staged as T(3) and eight (31 per cent) T(4). Eighteen (69 per cent) patients had N(0), seven (27 per cent) N(1) and one (four per cent) N(2) disease. Multiple primary tumours were recorded in three (11.5 per cent) patients. Four (15 per cent) patients had pre-operative radiotherapy with poor response and two (eight per cent) required emergency tracheotomy prior to surgery. Feeding jejunostomy was performed on 19 (73 per cent) and neck lymph node dissection in eight (31 per cent) patients. The mean duration of surgery was five hours (range 3.5 to 7.5) with a mean blood loss of 840 ml (range 160 to 1800), a mean stay in ICU of 4.2 days and hospital stay ranged from nine to 84 days (mean 34). Three (11.5 per cent) patients died (pneumonia - one, congestive heart failure - one, pulmonary embolus - one) in the early post-operative period. Eight (31 per cent) patients remain alive from 30 to 136 months (mean 58 months). Two (eight per cent) patients died with no evidence of disease. Thirteen (50 per cent) patients died of their disease between two to 51 months (mean 17.3 months) post-operatively. Kaplan-Meier survival estimates for one year was 65 per cent, for three years 35 per cent and for five years 26 per cent (see Figure 1). Median survival in the whole series was 18 months. Post-operative speech was with an electrolarynx in 16 (62 per cent). One patient (four per cent) used gastric speech and one patient (four per cent) used a Blom-Singer valve effectively. Five (19 per cent) patients had no speech post-operatively. All patients maintained oral feeding. Gastric transposition constitutes a safe and reliable method of restoring the continuity of the upper digestive tract following pharyngo-laryngo-oesophagectomy.

No abstract

The use of autoinflation in the treatment of glue ear has been widely recommended. In order to assess its efficacy, 40 children were studied. 21 children practised nasal autoinflation of a toy balloon 3 times daily for 3 weeks and 19 children had no treatment. In the autoinflation group only 4 glue ears improved, while in the untreated group 10 glue ears improved. This study suggests that autoinflation has no part to play in the treatment of glue ear in children.

Intractable epistaxis associated with septal perforation is a difficult problem to treat, particularly if nose-picking appears to be associated. Two such cases are presented in which Young's procedure of nasal closure was used to prevent epistaxis. Although the procedure was fully successful in only one patient, we feel the procedure has a part to play in the management of such cases of epistaxis.