Head and neck space infections present with a potential mortality rate of 40–50 per cent. This paper proposes an algorithm-based management of head and neck space infection to prevent life-threatening events.

A total of 225 patients with head and neck space infection were prospectively analysed at our institution. An experimental scoring system determined the level of clinical risk for the development of major complications. Accordingly, patients were classified into three risk groups: low-, intermediate- and high-risk.

Only intermediate- and high-risk patients were hospitalised. Intermediate-risk patients received intravenous medical therapy with daily re-evaluation; 18 of them required delayed surgery. Of the high-risk patients, three required immediate surgical treatment and five received delayed surgery, while in five cases medical therapy was the only treatment received. Low-risk patients were treated in an out-patient setting.

The algorithm-based management of head and neck space infection was successful in enabling the avoidance of lethal complications onset.

To assess the clinical utility of elective neck dissection in node-negative recurrent laryngeal carcinoma after curative radiotherapy for initial early glottic cancer.

A retrospective review was undertaken of 110 consecutive early glottic cancer patients who developed laryngeal recurrence after radiotherapy (34 recurrent T1, 36 recurrent T2, 29 recurrent T3 and 11 recurrent T4a) and received salvage laryngeal surgery between 1995 and 2005.

Six patients presented with laryngeal and neck recurrence and underwent salvage laryngectomy with therapeutic neck dissection, 97 patients with recurrent node-negative tumours underwent salvage laryngeal surgery without neck dissection and only 7 underwent elective neck dissection. No occult positive lymph nodes were documented in neck dissection specimens. During follow up, only three patients with neck failure were recorded, all in the group without neck dissection. There was no significant association between the irradiation field (larynx plus neck vs larynx) and the development of regional failure. A higher rate of post-operative pharyngocutaneous fistula development occurred in the neck dissection group than in the group without neck dissection (57.2 per cent vs 13.4 per cent, p = 0.01). Multivariate logistic regression analysis showed that early (recurrent tumour-positive, node-positive) or delayed (recurrent tumour-positive, node-negative) neck relapse was not significantly related to the stage of the initial tumour or the recurrent tumour. An age of less than 60 years was significantly associated with early neck failure (recurrent tumour-positive, node-positive).

Owing to the low occult neck disease rate and high post-operative fistula rate, elective neck dissection is not recommended for recurrent node-negative laryngeal tumours after radiation therapy if the initial tumour was an early glottic cancer.