Theories of post-laryngectomy anosmia

Stimulation of the olfactory neuroepithelium can occur orthonasally (i.e. due to odorant molecules entering the nose via the nostrils) or retronasally (i.e. due to delivery of odorant molecules via the nasopharynx and posterior choanaeReference Leon, Catalanotto and Werning⁷ to the olfactory cleft and the olfactory neuroepithelium). Most taste sensation depends upon retronasal stimulation of olfactory receptors; anosmia and hyposmia are therefore recognised to lead to reduced gustation and appetite.Reference Henkin and Smith¹²

There are two main theories regarding anosmia and hyposmia in patients who have undergone total laryngectomy.

The first theory states that reduction in nasal airflow leads to reduced olfaction due to the inability of odorant molecules to reach the olfactory mucosa.Reference Van Dam, Hilgers, Emsbroek, Touw, van As and de Jong^10, Reference Moore-Gillon^13–Reference Schwartz, Mozell and Youngentob¹⁶ Reduced nasal airflow is thought to be the cause of histological changes apparent in the olfactory neuroepithelium of laryngectomised patients, compared with normal subjects. Abnormalities include marked heterogeneity, frequent areas of degeneration, an absence of evidence of regenerative features, coverage of neuroepithelium by ‘pseudo-jaline’, replacement of the neuroepithelium by an abnormal alcianophilic material, and absence of neuroepithelium leaving exposed areas of tonaca propria on the nasal mucosa.Reference Minai, Ortolani, Bracale, Petrelli, Staffieri and Marchini⁸ Scanning electron micrographs of nasal mucosa from laryngectomised patients also consistently show a change from transitional-type epithelium to densely ciliated mucosa with an abundance of mucus.Reference Moore-Gillon¹³ Mucus production has been shown to reduce during the first 12 post-operative months, and the increased number of epithelial cilia have an increased ciliary beat frequency, although this does not translate into increased mucociliary clearance.Reference Fisher, Lund and Rutman¹⁷ However, when considering these findings it should be borne in mind that the majority of laryngectomy patients' olfactory mucosa will have had considerable exposure to tobacco smoke pre-operatively. Therefore, the effect of post-laryngectomy smoking cessation on the histological changes of the olfactory mucosa should not be disregarded.

The second theory of the aetiology of anosmia in laryngectomees was first proposed by Henkin et al. Reference Henkin, Hoye, Ketcham and Gould⁴ These authors postulated that laryngectomy caused damage to the sensory nerves of the larynx and ‘…significantly alter[ed] olfactory acuity through some complex feedback mechanism, or by interfering with a diffuse anatomical projection system from the pharynx and/or larynx to the olfactory brain’.Reference Ackerstaff, Hilgers, Aaronson and Balm² Basic science research has leant some support to this theory. Netzer et al. used a canine model to examine the olfactory neuroepithelium from two groups: (1) simulated laryngectomy (i.e. fashioning of an end-tracheostomy and denervation of the larynx); and (2) simple end-tracheostomy with retained laryngeal innervation. The histological changes found in the first group were very similar to those seen in laryngectomised patients.Reference Netzer, Goiz and Goldenberg¹⁸

The majority of authors lean towards the reduced nasal airflow theory as the primary cause of hyposmia and anosmia post-laryngectomy, although the argument developed in Henkin's paper, and Netzer's histopathological findings, cannot be ignored.

Interventions to improve olfactory performance

The methods reported to improve olfactory performance in laryngectomees all aim to improve nasal airflow as a means of delivering odorant molecules to the olfactory neuroepithelium. Two main techniques have been described: (1) use of a device to temporarily reconnect the mouth and nose to the lower airway; and (2) use of the oro-facial musculature to develop the means to generate nasal airflow.

Laryngeal bypass devices are well known and have been developed by physiciansReference Mozell, Hornung, Leopold and Youngentob^14, Reference Schwartz, Mozell and Youngentob^16, Reference Goktas, Lammert and Bert^19–Reference Tatchell, Lerman and Watt²² and patientsReference Verma, Verma and Jani²³ alike. Such devices use a flexible tube with a flange to create a seal around the stoma, with the other end held between the lips. Natural pulmonary inspiration therefore draws air in through the nose before passing it through the mouth, tubing and trachea. The patient must learn to relax the soft palate so it does not obstruct the nasopharynx. Studies have shown that patients who use this technique have increased olfactory accuracy, which diminishes when they do not.Reference Mozell, Hornung, Leopold and Youngentob^14, Reference Schwartz, Mozell and Youngentob^16, Reference Goktas, Lammert and Bert^19,Reference Tatchell, Lerman and Watt²² However, issues of practicality are frequently raised, and some patients find learning the required technique demanding.Reference Van Dam, Hilgers, Emsbroek, Touw, van As and de Jong^10, Reference Goktas, Lammert and Bert^{19, 20}

The technique whereby patients generate nasal airflow by a combination of facial and cervical muscle contractions has been noted as an innate attempt by the laryngectomee to improve olfaction.Reference Van Dam, Hilgers, Emsbroek, Touw, van As and de Jong^10, Reference Moore-Gillon¹³ It has also been noted that patients who develop good oesophageal speech may develop improved olfactory performance, due to the air swallowed into the oesophageal reservoir being inspired through the nasal cavity.Reference Gilchrist²⁴ These observations led to the development of the ‘polite yawning technique’ which ‘…consists of an extended yawning movement with simultaneous lowering of the jaw, floor of mouth, tongue, base of tongue, and soft palate while keeping the lips securely closed’.Reference Hilgers, Van Dam, Keyzers, Koster, van As and Muller⁵ This technique has been shown to be effective at converting laryngectomy patients considered ‘non-smellers’ into ‘smellers’, both in the short and long term, and can be effective after a single intervention, although further interventions lead to continued improvement in olfactory rehabilitation.Reference Hilgers, Van Dam, Keyzers, Koster, van As and Muller^5, Reference Risberg-Berlin, Moller and Finizia^9, Reference Risberg-Berlin, Yitalo and Finizia²⁵

A third technique, the use of a Rogers insufflator, was described in 1911 by Jobson Horne.Reference Gilchrist²⁴ It is interesting that the idea of capturing and forcefully blowing air into the nasal cavities has received little further attention in the literature since this publication.

Assessment tools

As with many methods of clinical assessment, standardisation of published results occurs rarely, and assessment of post-laryngectomy nasal function is no exception. A review of the different methods available to assess olfactory performance can be found elsewhere.Reference Simmen and Briner²⁶ A variety of methods have been described to assess olfactory performance in laryngectomised patients, including: subjective self-assessment of olfactory function;Reference Leon, Catalanotto and Werning^7, Reference Gilchrist²⁴ objective olfactory assessment using a variety of methods (Table I); chemosensory evoked olfactory potentials;Reference Welge-Luessen, Gerd and Wolfensberger¹¹ nasal airflow; mucociliary function; and speech therapist assessment of videos of patients performing the ‘polite yawn’.

Table I

Nasal function assessment methods used in laryngectomised patients

For objective assessment of olfaction, the University of Pennsylvania Smell Identification TestReference Doty, Shaman, Kimmelman and Dann³¹ is considered clinically as the ‘gold standard’, but its use has not been described in the laryngectomee. Most authors describe the use of a form of Odour Detection Test or Olfactory Discrimination Test.Reference Cain, Gent, Goodspeed and Leonard^27, Reference Hulshoff Pol, Hijman, Baaré and van Ree³² Odour Detection Tests establish the olfactory threshold by presenting a mixture of an odorant compound (e.g. phenyl-ethyl alcohol, a liquid with a floral odour) at different concentrations diluted in di-propylene glycol, an odourless substance,Reference Van Dam, Hilgers, Emsbroek, Touw, van As and de Jong¹⁰ or butanol (a highly alcoholic aroma) in deionised water. Decreasing concentrations of odorant compounds are usually used, which prevents olfactory fatigue and saturation.Reference Hilgers, Van Dam, Keyzers, Koster, van As and Muller⁵ Olfactory Discrimination Tests determine whether subjects can distinguish between two different smells without having to name them,Reference Van Dam, Hilgers, Emsbroek, Touw, van As and de Jong¹⁰ and often use common household items such as peanut butter, soap and chocolate.Reference Leon, Catalanotto and Werning⁷ Specific tests of retronasal olfaction have also been described.Reference Leon, Catalanotto and Werning⁷ These tests are time- and labour-intensive (the Olfactory Discrimination Test takes 30 minutes on average to complete). Therefore, it is of interest to note that Hilgers et al. used the Smell Disc Test as well as the Olfactory Discrimination Test combined with the Present Odour Perception Scale (see below) in their follow-up study of the use of the polite yawning technique in laryngectomees.Reference Hilgers, Jansen, Van As, Polak, Muller and van dam Frits⁶ The Smell Disc Test was found to be much simpler to use than the Olfactory Discrimination Test, and there was reasonable correlation between the Smell Disc Test score and the Olfactory Discrimination Test–Present Odour Perception Scale combination (Cohen's κ = 0.56; p < 0.001). They concluded that ‘[o]dour testing in individuals who have undergone a total laryngectomy is now possible in a relatively simple way using the easily applied [Smell Disc Test]’.Reference Hilgers, Jansen, Van As, Polak, Muller and van dam Frits⁶

Patients' own perception of their olfactory ability is another useful measure. The ‘Questionnaire concerning Olfaction, Taste and Appetite’ was first described by de Jong et al.,Reference de Jong, Mulder, de Graaf and van Staveren²⁹ and subsequently modified by Hilgers and colleagues.Reference Hilgers, Van Dam, Keyzers, Koster, van As and Muller^5, Reference Hilgers, Jansen, Van As, Polak, Muller and van dam Frits⁶ It comprises 31 multiple-choice questions relating to patients' present situation and their pre-operative experience. The questions are divided into five scales: present odour perception, odour perception now in comparison with the past, gustation, appetite and sensation of hunger. The Present Odour Perception Scale is part of the Questionnaire concerning Olfaction, Taste and Appetite, and has a scoring range of three to 15. Any patient scoring equal to or better than the mean score in the original reference groupReference de Jong, Mulder, de Graaf and van Staveren²⁹ can be considered to have normal olfactory acuity, i.e. a score of 10 or more.Reference Hilgers, Van Dam, Keyzers, Koster, van As and Muller⁵ The Questionnaire concerning Olfaction, Taste and Appetite appears to be a useful subjective assessment tool, although it has not yet been validated.

Chemosensory evoked olfactory potentials have been studied in laryngectomees by Welge-Luessen et al. Reference Welge-Luessen, Gerd and Wolfensberger¹¹ Of the 11 patients studied, evoked potentials were captured in eight when stimulated by hydrogen sulphide. No conclusions were drawn from this data.

Mucociliary function has been studied using saccharin clearance time, and has been shown to be significantly better in controls than laryngectomised patients (mean time ± standard deviation: 13.4 ± 1.25 minutes versus 18.0 ± 1.47 minutes, respectively (p < 0.01)).Reference Moore-Gillon¹³ The increased saccharin clearance time seen in laryngectomees may relate to the microscopic structural changes reported, i.e. an increase in the density of mucosal cilia and an increase in mucus production.

Nasal airflow measured with and without a larynx bypass device has also provided comparative data. Using a pneumotachometer, Mozell et al. showed that a laryngectomee with a larynx bypass device could produce nasal airflow tracings (measured at the nares) which were ‘remarkably similar’ to those of a normal subject.Reference Mozell, Hornung, Leopold and Youngentob¹⁴ Unsurprisingly, the use of a laryngeal bypass device produces significantly greater nasal airflow than if it is not used (27.4 l/min versus 5.4 l/min, respectively; p < 0.01),Reference Tatchell, Lerman and Watt²² and increasing maximal inspiratory sniff rate is positively correlated with olfactory ability, and significantly so if above 10 l/minute.Reference Moore-Gillon¹³ Only Henkin and Netzer's work tends not to support the theory that improved nasal airflow improves olfactory performance in the laryngectomee.Reference Henkin, Hoye, Ketcham and Gould^4, Reference Netzer, Goiz and Goldenberg¹⁸

Secondary outcome measures

It is widely stated that patients with olfactory problems have a relatively worse quality of life (QoL) than those without,Reference Leon, Catalanotto and Werning^7, Reference Miwa, Furukawa, Tsukatani, Costanzo, DiNardo and Reiter^33, Reference Hummel and Nordin³⁴ although it has not been established that this holds true for the laryngectomee. Two papers from the same group have used the European Organisation for the Research and Treatment of Cancer 30-item questionnaire, and the 35-item questionnaire from the same organisation specifically relating to patients with head and neck cancer, in order to evaluate the effect of loss of olfactory function on laryngectomy patients' QoL.Reference Risberg-Berlin, Moller and Finizia^9, Reference Risberg-Berlin, Yitalo and Finizia²⁵ These papers concluded that, when laryngectomised patients were categorised as either ‘smellers’ or ‘non-smellers’ according to the Scandinavian Odour Identification Test (see Table I), there was no significant difference in QoL between the two groups. In addition, these two papers found that interventions shown to improve olfactory performance also improved QoL scores, but not by a clinically significant amount (the improvement was −9.5 score points, whereas changes over time of 10 score points are considered clinically significant). In the latter of these papersReference Risberg-Berlin, Moller and Finizia⁹, this same group of authors stated that they ‘could not confirm the association between olfaction and quality of life when measured with the [European Organisation for the Research and Treatment of Cancer] questionnaires’.Reference Risberg-Berlin, Moller and Finizia⁹ They suggested that these findings may possibly be due to small cohort size, use of different assessment tools compared with other authors, and the fact that ‘…quality of life is a complex concept and depends on factors other than olfaction’.