Operating on the larynx of an awake patient is an old practice. Before general anaesthesia became safe and routine, laryngologists removed lesions under mirror guidance with cocaine as the only anaesthetic, and the indirect approach never entirely disappeared (Bastian and Delsupehe, 1996). Its return to the mainstream came with technology. Flexible endoscopes with a camera chip at the tip and a working channel gave a stable, magnified view that could be shared on a screen. Lasers that could be passed through that channel on a thin fibre, first the pulsed dye laser and then the potassium titanyl phosphate laser, allowed papilloma and dysplasia to be treated without suspension laryngoscopy (Zeitels et al., 2004; Zeitels et al., 2006). Injectable materials for vocal fold augmentation completed the set.

A substantial review a decade and a half ago already described office-based diagnosis and treatment as an established part of laryngology (Rosen et al., 2009), and later reviews have catalogued the procedures and their results (Shah and Johns, 2013; Wellenstein et al., 2018). What these sources assume is an experienced unit. A department starting from nothing has to make many decisions that the technical papers pass over: which procedures to offer first, in what room, with which staff, on which patients, and with what arrangements for the occasion when something goes wrong. This article proposes an ordered way of making those decisions. It describes a framework and reports no experience of applying it.

Three arguments are usually made. The first is avoidance of general anaesthesia, which matters most for the frail patients who make up a good share of those with vocal fold paralysis or recurrent papilloma. The second is functional: an awake patient can phonate, so the surgeon injecting a paralysed vocal fold hears the voice change and can judge the amount to place (Sulica et al., 2010). The third is cost and convenience. Comparisons of office and theatre pathways for injection laryngoplasty and for laser treatment of papilloma have found the office route considerably cheaper (Bové et al., 2007; Rees et al., 2007), and patients asked about unsedated laser procedures have mostly said they would choose the same again (Rees et al., 2006). Table 1 lists the procedures commonly moved to the office and the route by which each is done.

The case has limits that a new service should recognise from the outset. The view through a flexible endoscope is two-dimensional and the instruments passed down its channel are small, so the precision of microlaryngoscopy, with binocular magnification and two-handed dissection, is not reproduced. Biopsies taken through the channel are small and may under-represent a lesion. An equivocal or negative result from a lesion that looks suspicious should therefore lead to direct laryngoscopy and not to reassurance. Office laser treatment of papilloma controls the disease but generally has to be repeated at intervals. These are reasons to regard the office and the theatre as complementary, with a clear route from one to the other.

The reported complication rate of office laryngeal procedures is low, including in a large series of laser cases using three wavelengths (Koufman et al., 2007). Low is not zero, and the hazards differ from those of the operating theatre. The patient is conscious, unmonitored unless monitoring is arranged, and without a secured airway. Yung and Courey (2010) measured heart rate and blood pressure during office-based flexible endoscopic surgery and found rises large enough, in a notable share of patients, to qualify as severe hypertension or tachycardia, with no symptoms to alert the operator. Awake instrumentation of the pharynx can also provoke the opposite response, a vasovagal faint.

The airway itself may be compromised by laryngospasm, by bleeding, or by swelling after injection into an already narrow glottis. Topical anaesthesia removes the protective sensation of the larynx for a time, so that food and drink taken too soon may be aspirated. Local anaesthetic is applied to mucosa from which it is absorbed quickly, and several applications by different routes can add up to a large total. Lasers bring risks to the eyes of everyone in the room and are governed by formal standards (American National Standards Institute, 2018). A safe service is one in which each of these hazards has a named control, and Figure 1 pairs the stages of a procedure with the safeguard proposed for each.

The framework has eight components. They are described in the order in which a department would normally address them.

The framework is a proposal. Its elements are taken from published descriptions of office practice and from general patient safety measures, and each is reasonable on its face, but the package has not been introduced and evaluated in a department, and no claim is made that it lowers complication rates. Serious events in this setting are rare enough that demonstrating a reduction would require pooled data from many units. Several of its particulars are matters on which practice differs and evidence is thin: the place of routine blood pressure monitoring in patients without cardiovascular disease, the handling of anticoagulants, the length of observation, and how many supervised cases constitute adequate training.

Local conditions will also force changes. Regulations on lasers, on who may administer local anaesthetic and on the classification of procedure rooms differ between countries, and reimbursement rules can make an office service financially unattractive even where it is clinically sound. A unit adopting the framework should treat its first year as a pilot and publish what it finds.

Office laryngeal procedures under local anaesthesia offer patients a quicker and often safer alternative to the operating theatre, and the techniques are well described. Safety in this setting comes less from individual skill than from organisation: a defined scope, a prepared room, a team with distinct roles, selection criteria, a counted anaesthetic dose, monitoring, stop rules and audit. The framework proposed here gathers those elements into a form a department can adapt before it treats its first patient.

To be declared by the author(s) before publication.

To be declared by the author(s) before publication.