Clinical examination

Evaluation of the anosmic patient should always involve clinical assessment. The duration, speed of onset and pattern of olfactory disturbance should be determined (e.g. progressive, fluctuant). A thorough clinical history should be carried out, including presence of associated nasal symptoms, taste disturbance and allergy. A history of head injury is relevant, and details of the nature of the injury should be elicited, particularly regarding loss of consciousness, direction of impact and radiological findings.Reference Green, Rohling, Iverson and Gervais^6–Reference Callahan and Hinkebein⁸ A full medical history should be taken, including neurological, psychiatric and metabolic disorders (Tables I and II).Reference Seiden and Seiden^9–Reference Schiffman and Gatlin¹¹ Occupational history may reveal exposure to noxious chemicals (e.g. cadmium and benzene), and a smoking history is important.Reference Sugiyama, Matsuda, Kondo, Mitsuya, Hashiba and Murakami¹² Iatrogenic causes must be considered, including medication, neurosurgical intervention, radiotherapy and previous nasal surgery. Family history should be elicited. Some medications are implicated in altered smell and taste, including anticonvulsants and antihypertensives (Table III).Reference Doty and Bromley¹³

Table I Causes of olfactory loss

Reprinted with permission.Reference Seiden and Seiden⁹ URTI = upper respiratory tract infection

Table II Medical diseases causing olfactory dysfunction

ReprintedReference Jones and Rog¹⁰ from adapted material,Reference Schiffman and Gatlin¹¹ with permission. HIV = human immunodeficiency virus

Table III Medication causing olfactory dysfunction

ReprintedReference Jones and Rog¹⁰ from adapted material,Reference Schiffman and Gatlin¹¹ with permission.

The patient's general habitus may reveal clues as to the nature of the olfactory disorder. Congenital disorders of smell, including isolated absence or hypoplasia of the olfactory bulbs, are associated with Kallmann syndrome, Turner syndrome and premature baldness.Reference Singh, Grewal and Austin¹⁴ Nasendoscopy may show evidence of rhinosinusitis and polyposis, or reveal no abnormality. Cranial nerve examination should be included to assess for underlying neurological causes.

Olfactory testing

Cain describes three criteria necessary to maximise odour recognition in olfactory testing.Reference Cain¹⁵ Odours must be familiar to the patient, with a longstanding association between the odour and its name, and help should be given to recall the name. Reliability is improved using both threshold testing and odour discrimination assessment. ‘Forced-choice’ procedures reduce response bias. Patients scoring less than chance are likely to be malingering, as are those who fail to identify trigeminal nerve stimulants such as ammonia or 4 per cent butanol.

The University of Pennsylvania smell identification test (UPSIT) system was most commonly used by UK respondents to our questionnaire. This is a forced choice test, with 40 microencapsulated odours, acting as a ‘scratch and sniff’ test. This test can indicate a level of olfactory function (i.e. mild to total anosmia) and has a score ranking for age and gender.Reference Doty¹⁶ However, the UPSIT system has not been validated in a UK population, relies on supra-threshold testing and is relatively expensive. The Cross-Cultural Smell Identification test is a12-item test based on the UPSIT, which can be carried out in five minutes and is self-administered.Reference Doty, Marcus and Lee¹⁷ ‘Sniffin’ Sticks' is a test of olfactory function based on felt-tip pens and assesses odour threshold, discrimination and identification.Reference Kobal, Hummel, Sekinger, Barz, Roscher and Wolf¹⁸ Doty et al. evaluated 10 different olfactory tests, noting that there was considerable variation in their reliability, related to the length of the tests. They suggested that the results from different testing methods should not be compared, as variations may be a result of differing reliabilities rather than reflecting clinical findings.Reference Doty, McKeown, Lee and Shaman¹⁹

Currently, the only test validated in a UK population is the Combined Olfactory Score.Reference Robson, Woollons, Ryan, Horrocks, Williams and Dawes²⁰ The suprathreshold test consists of nine odours and ammonia, while the threshold test uses dilutions of 1-butanol from 0.00061 per cent to 4 per cent in nine steps. The scores are combined and the average of the two is the final score. The test also allows differentiation between severities of olfactory dysfunction.

As well as a diagnostic and quantitative role, formal olfactory testing allows monitoring of progression or resolution of dysosmia, particularly following surgical or other therapeutic intervention.

Radiological evaluation of olfactory dysfunction

Imaging may be required in the evaluation of olfactory disorders. Computerised tomography is the most appropriate technique for patients with sinonasal disease, regarding surgical planning. However, magnetic resonance imaging is more useful for diagnosis of olfactory apparatus abnormalities and parenchymal disease, particularly in congenital disease. Absent olfactory bulbs and hypoplastic olfactory sulci are particularly noted in Kallmann syndrome, along with loss of temporal and/or frontal lobe volume.Reference Madan, Sawlani, Gupta and Phadke^21, Reference Yousem, Oguz and Li²² Decreased volume of the olfactory bulbs is noted with increasing age.

Patients with olfactory groove or frontal lobe meningiomas are likely to present with hyposmia.Reference Welge-Luessen, Temmel, Quint, Moll, Wolf and Hummel²³ These may reach a significant size (>4 cm) before presentation, due to a gradual deterioration in olfactory function and preservation of unilateral olfactory function.Reference Welge-Luessen, Temmel, Quint, Moll, Wolf and Hummel²³ Welge-Leussen et al. suggest that all patients with lateralised dysosmia should undergo radiological evaluation, after noting that 50 per cent of patients with olfactory meningiomas had unilateral dysosmia on formal testing. Recovery of olfaction following intracranial surgery is not well documented, although Ishimaru et al. report improved olfactory function in a patient following decompression of a right frontal lobe meningioma.Reference Ishimaru, Miwa, Nomura, Iwato and Furukawa²⁴

Accurate diagnosis of skull fracture site and associated parenchymal injuries may allow prediction of the likelihood of recovery of smell.Reference Ogawa and Rutka⁷ A positive correlation has been shown between number of plaques and olfactory function in patients with multiple sclerosis. Functional assessment of hyposmia with single photon emission computed tomography (SPECT) imaging demonstrates reduced frontal blood flow in patients with schizophrenia.Reference Yousem, Oguz and Li²²

Although multiple pathologies can be demonstrated on neuroradiological assessment, there are no guidelines regarding the indications for imaging, particularly relating to the sensitivity and specificity of imaging techniques. In our survey, 81 diagnoses of anterior cranial fossa tumours were made by 51 respondents, one surgeon reporting four cases alone. However, a number of respondents commented that they had never seen a single case, in a long established career. Busaba studied 28 patients with anosmia and negative endoscopy results, finding that imaging did not add any further information and concluding that it should not be requested routinely.Reference Busaba²⁵ This appears to be the only study evaluating this aspect of radiological assessment. However, it was a small retrospective, unblinded study of 20 patients and the conclusions should be interpreted with caution.

Olfactory dysfunction following head injury

Head injuries account for 18 per cent of olfactory disturbances.Reference Deems, Doty, Settle, Moore-Gillon, Shaman and Mester⁵ Olfactory insult results from damage to nasal mucosa, shearing of olfactory fibres due to cribriform plate fracture, and oedema of the olfactory tracts and bulbs. Damage to the peripheral olfactory apparatus results in anosmia, whereas central olfactory damage manifests as an inability to discriminate odours.Reference Leopold, Schwob, Youngentob, Hornung, Wright and Mozell^26, Reference Leopold and Myerrose²⁷ The anterior temporal lobes and orbitofrontal poles are most vulnerable to damage, where reduced orbitofrontal perfusion is demonstrated on SPECT.Reference Reiter, DiNardo and Costanzo^28, Reference Varney and Bushnell²⁹ The severity of post-traumatic olfactory impairment is more pronounced, with less chance of recovery than in cases of infection or chronic rhinosinusitis.

Potential predictive factors may allow identification of the likelihood of recovery. Green et al. retrospectively analysed 367 patients with post-traumatic olfactory disorders, finding a sharp decrease in olfactory function proportional to injury severity.Reference Green, Rohling, Iverson and Gervais⁶ Patients with a Glasgow Coma Scale score of less than 13 at presentation, post-traumatic amnesia and radiological abnormalities were markedly less likely to recover their sense of smell. Ogawa and Rutka noted a similar relationship, with a reduced chance of smell recovery following loss of consciousness of more than one hour and occipital, frontal and skull base fractures.Reference Ogawa and Rutka⁷ Callahan and Hinkebein noted that 40 per cent of such patients suffered an olfactory deficit, which only manifested on formal testing.Reference Callahan and Hinkebein⁸

Recovery of a normal sense of smell following a head injury is unlikely, although improvement can occur over a longer time period that previously realised. Recovery has been noted up to 18 months post-injury, whereas Duncan and Seiden demonstrated improved olfactory scoring up to five years post-injury, with 35 per cent of patients improving in total.Reference Reiter, DiNardo and Costanzo^28, Reference Duncan and Seiden³⁰ Doty et al. evaluated 268 patients between one and 13 years following a head injury;Reference Doty, Yousem, Pham, Kreshak, Geckle and Lee³¹ no patient with post-traumatic anosmia returned to normosmia, but patients' olfaction improved over time in 36 per cent, particularly in those with parosmia (dropping from 41 to 15 per cent over eight years). Recovery, again, was proportional to severity of head injury.

Olfactory dysfunction following upper respiratory tract infection

Temporary anosmia can occur with an upper respiratory tract infection, when oedema prevents odorant molecules reaching the olfactory cleft. Viral upper respiratory tract infection accounts for 20–30 per cent of identified cases of olfactory loss,Reference Seiden³² typically caused by the parainfluenza 3 virus.Reference Sugiura, Aiba, Mori and Nakai³³ In a small percentage, olfaction remains permanently distorted, particularly in women (70–80 per cent) and in those aged 40–60 years.Reference Mott and Leopold³⁴ This is partly due to cumulative degeneration of the olfactory apparatus with age.Reference Doty¹⁶ It is thought that viral infections cause a reduction in the number of olfactory receptors, with replacement by respiratory epithelium.Reference Seiden^32, Reference Jafek, Murrow, Michaels, Restrepo and Linschoten³⁵ However, stem cells with the potential for regeneration may persist.Reference Hummel³⁶ The reported prognosis for upper respiratory tract infection induced hyposmia varies. Hummel found that the majority of patients recovered function within six months, whereas Cullen and Leopold noted that some patients continued to recover up to three years following the initial olfactory insult.Reference Hummel^36, Reference Cullen and Leopold³⁷ Duncan and Seiden found that 19 of 21 patients with viral-induced hyposmia had markedly improved UPSIT scores at three years, although Doty and Mishra documented minimal recovery in such patients.Reference Duncan and Seiden^30, Reference Doty and Mishra³⁸

Olfactory dysfunction after rhinosinusitis

If olfactory dysfunction secondary to nasal disease is due to odorant molecules failing to reach the olfactory apparatus, it seems logical that treating mucosal oedema and polyposis would result in a symptomatic improvement.

The effectiveness of surgical intervention in treating hyposmia secondary to chronic rhinosinusitis is open to debate (Table IV). A small study by Damm et al. found a significant correlation between nasal airflow and odour identification in patients with chronic rhinosinusitis, suggesting that surgery to improve nasal airflow and eliminate mucosal disease would be helpful.Reference Damm, Eckel, Streppel, Jungehiilsing and Stennert³⁹ A non-blinded, retrospective study by Iro et al. reported the success of endoscopic sinus surgery (ESS) in reducing nasal symptoms, including anosmia.Reference Iro, Mayr, Wallisch, Schick and Wigand⁴⁰ They reported a 92 per cent success rate over three years, although this was based purely on subjective data. Rowe-Jones and Mackay collected prospective data on 115 patients, evaluating subjective symptoms and olfactory detection thresholds, prior to and six weeks following ESS.Reference Rowe-Jones and Mackay⁴¹ All parameters significantly improved, including nasal volume on acoustic rhinometry. Improvement in olfactory scores was proportional to the increase in nasal volume. However, Kimmelman found no improvement in olfaction scores in patients undergoing nasal polypectomy.Reference Kimmelman⁴²

Table IV Levels of evidence for medical and surgical management of olfactory disorders

UPSIT = University of Pennsylvania smell identification test; incl = including; VAS = visual analogue scale; vit = vitamin

Landis et al. note that, although 76 per cent of post-ESS patients improved their olfactory function, 13 per cent demonstrated a deterioration on formal testing.Reference Landis, Giger, Bouayed, Hugentobler, Plouin-Gaudon and LaCroix⁴³ In an extensive review of the literature, Deems et al. comment that both surgical and medical interventions do not result in return to normosmia.Reference Deems, Doty, Settle, Moore-Gillon, Shaman and Mester⁵ Similarly, the relationship between airway patency and olfactory function is questionable.Reference Doty and Mishra³⁸ This could be explained by the findings of Lee et al., who noted lower levels of olfactory epithelium and replacement with normal respiratory mucosa in patients with chronic rhinosinusitis.Reference Lee, Lim, Lee, Park and Choi⁴⁴ Inflammatory changes within the olfactory mucosa may account for hyposmia, independent of airflow alteration.Reference Kern⁴⁵ The likelihood of recovery of olfaction in patients with chronic rhinosinusitis or polyposis seems to be time-dependent, with prolonged disease resulting in degeneration of olfactory mucosa and persistent olfactory dysfunction.

Pharmacological therapy for olfactory disorders

Clinical experience often suggests that a patient's sense of smell may return following treatment with corticosteroids. Hotchkiss described a subjective improvement in olfaction in patients with chronic rhinosinusitis treated with 70 mg prednisolone.Reference Hotchkiss⁴⁶ However, no formal olfaction testing was carried out as confirmation. Jafek et al. describe two patients in whom high dose steroids temporarily restored a normal sense of smell, prior to surgery for nasal disease, followed by long-term, low dose nasal steroids.Reference Jafek, Moran, Eller, Rowley and Jafek⁴⁷ These authors felt that a trial of steroids was worthwhile in patients with allergic rhinitis, nasal polyposis and anosmia (Table IV). In the study by Tos et al., steroid nasal sprays were given to hyposmic patients with rhinosinusitis.Reference Tos, Svendstrup, Arndnal, Orntoft, Jakobson and Borum⁴⁸ Olfaction scores improved but significant persistent hyposmia was noted in a number of subjects. Similarly, Mott et al. noted an improvement to the mid-hyposmic range in patients treated with flusenolide, although this was an unblinded trial without placebo control.Reference Mott, Cain, Lafreniere, Leonard, Gent and Frank⁴⁹

Golding-Wood et al. performed UPSIT scoring on patients with perennial rhinitis treated with intranasal betamethasone.Reference Golding-Wood, Holmstrom, Darby, Scadding and Lund⁵⁰ Patients with scores in the hyposmic range improved but failed to reach normosmia. There was no improvement in anosmic patients. Blomqvist et al. describe a significant improvement in olfactory thresholds in 48 consecutive patients with anosmia or hyposmia receiving a 10-day course of oral prednisolone (40 mg for 3 days, followed by a daily reduction of 5 mg) and fluticasone nasal spray.Reference Blomqvist, Lundblad, Bergstedt and Stjarne⁵¹ However, the improvement in olfactory scores was maintained equally well whether patients continued with nasal fluticasone or placebo, when compared with controls. These authors felt that the improvement was probably due to a reduction in mucosal oedema, even though a conductive olfactory loss may not be apparent. Heilmann et al. were unable to demonstrate an improvement in olfaction following use of topical mometasone nasal spray in patients with sinonasal disease or upper respiratory tract infection related hyposmia, but they noted a significant increase in olfactory scoring following treatment with systemic corticosteroids.Reference Heilmann, Huettenbrink and Hummel⁵² Similar findings were noted by Stevens; patients who remained anosmic following ESS or polypectomy were found to respond well to oral corticosteroids but not to topical application.Reference Stevens⁵³

The evidence suggests that a trial of oral corticosteroids may be useful, although there is little information on the required dose or length of treatment. However, long-term use of corticosteroids is likely to have undesirable side effects which outweigh the handicap of hyposmia.

A small number of studies have assessed substances other than steroids for the treatment of olfaction disorders. Zinc deficiency has been postulated as a contributing factor in hyposmia. Henkin et al. found no therapeutic effect in a randomised, placebo-controlled trial of zinc sulphate in hyposmic patients; however, Aiba et al. described significant improvement in patients with post-traumatic hyposmia, following zinc replacement.Reference Aiba, Sugiura, Mori, Matsumoto, Tomiyama and Okuda^54, Reference Henkin, Schecter, Friedewald, Demets and Raff⁵⁵ This appears to be based on retrospective, unblinded evidence. Jafek et al. reported that topical application of zinc gluconate resulted in permanent destruction of the olfactory epithelium.Reference Jafek, Linschoten and Murrow⁵⁸

A significant improvement in olfaction was described by Hummel et al. in patients treated with alpha-lipoeic acid following upper respiratory tract infection.Reference Hummel, Heilmann and Huttenbriuk⁵⁶ This trial involved 23 patients and was unblinded, with no crossover limb, and should therefore be interpreted with caution. Finally, Quint et al. evaluated the efficacy of caroverine in the management of non-conductive olfactory disorders.Reference Quint, Temmel, Hummel and Ehrenberger⁵⁷ Caroverine is thought to act by preventing glutamatergic neurotoxicity and was compared to the effects of oral zinc sulphate as a control. Quint et al. found a significant improvement in odour thresholds and identification in the treatment group, with no improvement in the zinc sulphate group. However, the study does not appear to have been blinded or randomised.

The evidence therefore suggests that, other than corticosteroids, there are no other successful pharmacological treatments for anosmia.