Introduction

ANOSMIA comes from the Greek; an [no] -osmia [smell]. This website is devoted to the loss and disturbance of the sense of smell and is intended as a resource for sufferers from olfactory disorders. The information it contains represents a review of the available literature and does not constitute medical advice. If you need medical advice then you should consult a physician (US) or your general practitioner (UK).

To many people, the ability to sense thousands of different odours is a normal occurrence and something that they may take for granted. However, to a minority, this natural skill is lost due to a disorder of the olfactory system. For them the enjoyment of smelling pleasant odours such as freshly cut grass is prohibited. During human evolution, having the ability to smell has been fundamental for survival, enabling us to identify healthy food sources and detect danger. Nowadays, however, this skill is not so necessary in developed countries due to their abundant food supplies. Nevertheless, the loss of the sense of smell can have profound psychological and somatic consequences (Van Toller, 1999) - a fact that is little recognised by the general public or medical profession.

Odours can bring both enhancement and improvement of life for an individual, whether it be the smell of food, fragrances or associating a particular odour with an event or person - we all know the smell of our loved ones. More than this, smell has other functions. Taste is to a great extent determined by smell (try eating while holding your nose), so a loss of smell often leads to a loss of taste. Smell is involved in relaying emotions; for example it can signal emotions such as fear and anxiety, and may be involved in other less understood aspects of human communication and sex.

Distortions to olfactory sensation can therefore cause great disturbance to our lives. There is a loss of quality of life and it can bring anxiety and loss of appetite. It can lead to depression. The inability to detect smoke can be dangerous and lead to subsequent harm. Food poisoning is more prevalent in patients who cannot detect rotten food.

Due to the wide diversity of olfactory disorders, some cases maybe permanent whilst others only temporary, depending on the cause and patient history. Some abnormalities in smell can signal the existence of several diseases of the nervous system such as Parkinson's and Alzheimer's Disease. Currently knowledge and understanding about the sense of smell is limited, although the past two decades has seen some encouraging results.

 

Definitions and Developmental Onset

It has been estimated that two million Americans suffer with smell and taste disorders and about 200,000 visit a doctor each year (Jafek et al., 2000; Seiden, 1997; Duncan and Smith, 1995; Smith and Seiden, 1991). Nevertheless, because this disorder is not life threatening, its importance is frequently underestimated. Any information about diagnosis is currently limited and doctors have difficulty prescribing a method of treatment.

A. Definitions

There are many definitions of the various categories of smell dysfunction. The following definitions come from Seiden (1997):

Olfactory disorders
ANOSMIA a complete loss of smell
HYPOSMIA partial loss of smell
HYPEROSMIA enhanced smell sensitivity
DYSOSMIA distortion in odour perception (includes parosmia & phantosmia)
PAROSMIA distortion of perception of external stimulus
PHANTOSMIA smell perception with no external stimulus

 

Generally these are the supported views of several scientists, but in an earlier publication Smith and Seiden (1991) define hyposmia as: a decreased sensitivity to some or all odorants. Leopold (1995) believes that there are three main types of disorder; anosmia and hyposmia ( as mentioned above), but also troposmia. This is a term that Leopold devised from the prefix "trop" meaning to turn or react, and "-osmia" meaning sense of smell. Troposmia is associated with an odorant stimulus that sufferers perceive to smell foul or rotten when in fact it is a pleasant odour.

Troposmia is thought to be associated with a reduced olfaction ability that could be linked to some cancers or may occur after an upper respiratory infection (URI), head trauma or after treatment with antibiotics (Leopold, 1995). There are reports that those patients who do lose their sense of smell after neural loss can regain their olfaction ability. Sometimes the distorted smell can be masked so that everything smells rotten and unpleasant. This can occur at varying levels, for example each odorant has a unique smell but each odour has the same slight distortion. Indeed, in some cases the true odour is completely replaced with a 'new' unpleasant substitute. This makes clinically testing these patients even more difficult because they have to be instructed to answer using two categories, the 'remembered' smell or the 'new' smell (Leopold, 1995).

Currently dysosmia can not be measured using modern techniques. As a disorder it is thought to occur in combination with olfactory loss such as that induced by infection. The distortions may be associated with intact but impaired olfactory system, whilst phantom smells may occur due to a disconnected system. It has been suggested that dysosmia is a sign of regeneration and eventual recovery, but further investigation into this idea does not support this view (Seiden, 1997). However, Smith et al. (1987) did find that it is more common in patients with post-traumatic hyposmia than anosmia (cited from Seiden, 1997). According to Seiden (1997), dysosmia is believed to be either intermittent or constant but almost always unpleasant. Some sufferers have described smells as "sewer-like" or "rancid".

 Patients with phantosmia complain that they can perceive a foul smell that no one else can, hence the word 'phantom'. When diagnosing a patient, a distinguishable characteristic is that the odour is usually perceived by one nostril. When writing about his own experiences with sufferers of phantosmia, Leopold (1995), explained the onset of the disorder. Initially, the unpleasant odour is perceived for only a few minutes and the patient is often amazed that no one else is experiencing the same smell. Then the second encounter may occur several weeks later showing both the same characteristics and duration. Gradually the symptoms become more frequent, until one or two years after the initial experience, daily episodes are reported that last for most of the day. Patients are usually unaware that the perceived foul is only produced from one nostril. Leopold (1995), also suggests that it is often associated with sudden changes in nasal airflow such as sneezing or snorting.

B. Patient Diagnosis

There are several methods of diagnosis used by physicians with respect to olfaction disorders. The information obtained can pinpoint the exact aetiology (Feldman et al., 1986).

 i) Patient's History

A thorough patient history and physical evaluation are essential for establishing the cause of the olfactory disorder (Jafek et al., 2000; Cullen and Leopold, 1999; Seiden, 1997, Leopold, 1995). In particular, when investigating the patient's history, identifying the time that the loss occurred and whether any other conditions may have contributed, are fundamental for a diagnosis (Jafek et al, 2000).

It is important to distinguish between a smell related disorder and a taste related disorder because they are commonly misused or misunderstood (Cullen and Leopold, 1999). Some patients who report an olfactory disorder, actually have a taste complaint and vice versa. There are a few cases where patients have shown evidence of both disorders, but generally it is more likely to be related to one (Leopold, 1995).

For a complete diagnosis, quantitative measurements of the patient's olfactory system need to be obtained (Smith and Seiden, 1991). As summarised by Doty (1997) quantitative olfactory assessment is essential for several reasons:

a) to establish the validity of a patient's complaint

b) characterise the specific nature of the problem

c) reliably assess the effectiveness of treatments

d) detect malingering

e) establish compensation for disability

 

It is important to identify events prior to the onset of the disorder and some of the questions asked are the length of time of olfactory loss and whether it was an intermittent or continuous olfactory loss (Cullen and Leopold, 1999). Fluctuation is usually a sign of conductive loss and hence may indicate a nasal or sinus disorder (Seiden, 1997). Patients are often asked to describe their abnormal or distorted perception because this enhances the diagnosis. In most cases a family history is not needed, except when there is an apparent genetic loss. However, when a toxin or environmental factor is thought to be responsible then obtaining a social history may be important.

ii) Physical Examination

It is widely believed that when evaluating a patient a complete head and neck examination is essential (Jafek et al, 2000, Cullen and Leopold, 1999, Seiden 1997). A full physical examination usually takes place after a physician's referral. The examination will focus on the nose to determine whether it is a conductive or sensorineural loss. Use of an endoscope provides reliable observations of the olfactory cleft and epithelium, in particular the condition of the lining and the amount of mucous. It also identifies any obstructions within the olfactory system. Anterior rhinoscopy using an otoscope enables detection of nasal polyps or inflammation of the neuronal membrane. Examination of the neurological system looks at the cranial nerves, cerebellar functions and sensorimotor functions (Cullen and Leopold, 1999).

iii) Chemosensory Testing

According to Smith and Seiden (1991), assessment of a patient's disorder must involve a well-designed and standardised sensory test, that incorporates a controlled stimulus with a clinical setting. Over the years, several tests have been developed, some more successful and reliable then others. With regards to controlling the concentration of an odour, several approaches have been designed, for example 'sniff' bottles and injecting odorants directly in the nostrils. The main aim of these sensory tests is to assess the degree of dysfunction (Smith and Seiden, 1991).

 In the early 1980s, Doty et al., developed the University of Pennsylvania Smell Identification Test (UPSIT) (Smell Test site) . Currently, this is the most favoured sensory test in the U.S. (Jafek et al, 2000). It involves identifying forty microencapulated odours from a four-choice, forced-choice selection of possible odours and is available commercially (Archer, 2000, Jafek et al, 2000). As a test, it meets all essential criteria needed because it is both practical and clinical (Doty, 1997). It is also very popular because it can be self-administered, lasting ten to fifteen minutes and then scored by non-medical personnel within a minute (Doty, 1997). To assess the type of dysfunction, an individual's score is compared with data from a control of the same age and gender. As a general rule a normal score is between 30-40 (with some age considerations), whilst a hyposmic scores in the 20s and an anosmic falls close to 10 (Jafek et al, 2000). It is widely believed that this is a reliable method of diagnosing olfactory disorders. Assessing the reliability of some tests will be discussed in a later chapter. The UPSIT test has proven both reliable and reproducible as a simple clinical test of olfaction (Golding-Wood et al., 1996).

iv) Neuroradiologic evaluation

The main causes of smell deficit can be diagnosed after obtaining the history and physical examination, but for those patients whose problem is described as 'idiopathic', then additional examining using magnetic resonance imaging (MRI) or computer tomography (CT) scans may be needed. These images can then identify any abnormalities within the central nervous system and in the case of MRI, can detect any tumours within the olfactory system (Archer, 2000). If a CT scan shows that the olfactory cleft is open then nonreversible impairment is confirmed, whereas if it obstructed then another cause such as inflammation, is considered (Davidson et al., 1995). It is generally accepted that MRI scans are the radiological study of choice especially in the workup of anosmia (Archer 2000). However, according to Davidson et al (1995), MRI is not appropriate when compared to CT because, as they claim, MRI overemphasises swellings within soft tissue and poorly defines the bony detail of the cribriform and cleft areas. They insist that a patient who has had an MRI should also have a CT scan.

Causes of Olfactory Disorders

It is commonly believed that olfactory dysfunction is associated with a number of systemic and metabolic disorders. However, those patients who suffer from a primary smell complaint usually fall into a one of four main categories

  1. nasal and/or sinus disease, i.e. nasal polyposis, chronic sinusitis
  2. prior upper respiratory infection, i.e. a history of a viral-like upper respiratory illness just prior to the onset of olfactory loss
  3. idiopathic
  4. head trauma (Smith and Seiden, 1991)

In an experiment by Seiden (1997), 339 patients were diagnosed at the University of Cincinnati Taste and Smell Centre, and of these 84% had either olfactory loss or dysosmia.

 

Table 1. CAUSES of OLFACTORY LOSS
Aetiology % patients
Head injury 19*
Post URI 17*
Nasal/sinus disease 16*
Idiopathic-nasal 17
Toxic exposure-nasal 5
Multiple 5
Congenital 2
Age 1
Idiopathic-oral 9
Miscellaeous-oral 6
Toxic exposure-oral 1

 * Main causes of olfactory loss, which accounted for >50% of the reported cases. (Original data obtained from Seiden, 1997).

URI = upper respiratory infection

In most studies there is a common belief that the largest cause of olfactory dysfunction was either nasal and/or sinus disease or URI. Any fluctuations in results may reflect the nature of the referral patterns within the Clinics (Smith and Seiden, 1991).

i) Nasal and/or Sinus Disease (NSD)

There are believed to be two main categories of olfactory dysfunction, disorders of conduction and sensorineural injuries (Smith and Seiden, 1991). In conduction injuries, an odour cannot gain contact with the functioning olfactory system, whereas in sensorineural injuries, an odour does make a connection, but processing a response is impeded. From the four main aetiologic categories, nasal and/or sinus disease (NSD) is the only one that typically involves a conductive loss (Smith and Seiden, 1991).

When an olfactory disorder is generated from NSD it is usually quite severe and most patients are subsequently diagnosed as anosmic (75%) rather than hyposmic (Smith and Seiden, 1991). Obstruction of the nasal vault may arise due to intranasal polyposis (polyps), chronic sinusitis or allergic rhinitis. If this is thought to be the problem then it is important that the patient has a full endoscopic evaluation of the nasal cavity. As explained by Duncan and Smith (1995), prior to the results of endoscopic or radiological examination, the patient's history may contain information suggesting a diagnosis of NSD. When looking at the patient's history, such events such as a gradual onset of symptoms or fluctuations in smell sensitivity are more likely to suggest NSD than any of the other aetiologies. Some patients do report that their sense of smell does return temporarily after exercise or with medications (Duncan and Smith, 1995; Smith and Seiden, 1991), but this will be explained in more detail in the section on treatments (see below). If a patient complains of allergic rhinitis, often due to IgE-mediated nasal allergy, the deficit may be worse than in patients with nasal polyps.

ii) Head trauma/ Post traumatic injuries

This is believed to be the highest cause of anosmia. Patients are usually younger, between 20 and 50 years, and male, because they are more likely to suffer head trauma (Jafek et al., 2000; Seiden, 1997). The regions of the brain that are mainly affected are the frontal and occipital lobes. The degree of olfactory impairment is related to the extent and site of injury (Jafek et al., 2000; Duncan and Smith, 1995; Duncan and Seiden, 1995). Indeed, head trauma can cause detachment of the olfactory filaments from their location at the top of the cribriform plate which prevents olfactory input signalling to the olfactory bulb (Deems et al., 1991). Changes to the neuroepithelium suggests degeneration of a severe olfactory nerve. Several animal studies have provided evidence of intracranial haemorrhage and cerebral ischemia that have resulted degeneration of the olfactory epithelium despite the olfactory nerve remaining intact (Duncan and Smith, 1995; Smith and Seiden, 1991). Therefore, it is conceivable that a number of mechanisms may be involved in olfactory dysfunction following head trauma (Duncan and Smith, 1995; Smith and Seiden, 1991).

Loss of smell sensitivity is usually accompanied by dysosmia and many patients remark on their unpleasant smell sensations. As with nasal/ sinus aetiology, olfactory impairment is likely to be severe and therefore almost 80% report symptoms of anosmia rather than hyposmia (Smith and Seiden, 1991).

iii) Prior Upper Respiratory Infection (URI)

This aetiology accounts for 20-30% of all olfactory losses (Cullen and Leopold, 1999). It is highest in women (Cullen and Leopold, 1999; Seiden,1997; Duncan and Smith, 1995), and is mainly associated with hyposmia and as many as two-thirds may complain of dysosmia (Jafek et al., 2000). It is widely believed that this olfactory disorder is harder to manage because it involves sensory olfactory loss (Duncan and Smith, 1995, Smith and Seiden, 1991). A diagnosis can be made if olfactory dysfunction occurs immediately after an upper respiratory infection.

 In 1989, Jafek and Eller, carried out an ultrastructural study of the human olfactory epithelium (cited from Smith and Seiden, 1991). Their results showed a loss in the number of olfactory receptor cells and complete absence of olfactory cilia on the remaining receptors cells in patients of URI. A similar result was reported and summarised in a review by Cullen and Leopold (1999), where data from a biopsy showed that cilia were missing from the olfactory dendrites and there were fewer olfactory neurones and axons in URI patients. This would all seem to correlate with the known fact that influenza viruses can affect ciliary activity in the respiratory epithelium.

Any long-lasting effects from a viral illness may be the result of damage to the olfactory epithelium or could even involve central olfactory pathways (Smith and Seiden, 1991). There may be evidence of retrograde degeneration on the olfactory receptor cells in the olfactory bulb (Duncan and Smith, 1995). Indeed, one or all of these cases could result in either hyposmia or anosmia. However, if the reduction in olfactory receptors is accountable for URI, then it is important to remember that these patients are usually elderly and have fewer olfactory receptors naturally due to ageing (Duncan and Smith, 1995).

iv) Idiopathic

Within this category are patients who suffer from a primary olfactory disorder but not one that can be easily assigned to the other aetiologies. Their complaint may be due to a number systemic diseases, neurologic or endocrine disorders. As our understanding of olfactory disorders improves, the number of cases ascribed to this aetiology will probably decrease.

Although almost 80% of disorders will fall into one of the four categories (Smith and Seiden, 1991), the remaining 20% will be diagnosed from other causes. For a few patients over-exposure to toxins and environmental agents such as chloride gases and solvents, has led to their disorder. Whilst for others, factors such as malnutrition have implicated their smell dysfunction and some doctors believe that vitamin or trace metal imbalances are responsible neuronesible (Smith and Seiden, 1991). Currently, there is much debate about treatment involving zinc (the 'zinc controversy') and this will be discussed below in the section on treatment. About 5% of patients have multiple causes, and this usually occurs when an initial diagnosis has been made but then a second factor can not be ruled out (Seiden, 1997).

v) Central Nervous System Disorders

It is generally known that a number of CNS disorders are associated with olfaction dysfunction, in particular phantosmias have been associated with seizure activity and even Alzheimer's Disease (Seiden, 1997). If these smell sensations are brief then in some cases loss of consciousness and electroencephalographic (EEG) readings would show that they demonstrated either temporal or uncal seizures. Patients with temporal lobe epilepsy show signs of deficiency in odour identification and odour recognition (Smith and Seiden, 1991).

 Loss of smell sensitivity is one of the first signs of Alzheimer's disease (Doty, 1997) and this is believed to be associated with deficits in both odour identification and threshold (Smith and Seiden, 1991). Evidence has shown signs of neural plaques, neurofibrillary tangles and cell loss in the anterior olfactory nucleus (Thompson et al., 1998; Li et al, 1995) and throughout other olfactory-related regions of the brain (Smith and Seiden, 1991). It is also thought that olfaction deficits occur in these patients because the olfactory bulb has extensive connections with the areas of the brain that are affected by Alzheimer's (Thompson et al., 1998).

Currently, it is unclear whether olfactory dysfunction seen in Alzheimer's is the same for Parkinson's disease. It is known that Parkinson's is caused by the depletion of dopamine levels in the brain. However, whether the olfactory deficit is related to this is debatable because L-dopa therapy does not improve smell sensitivity, hence more research is needed (Li et al., 1995). 

With regards to schizophrenia, there is now clear evidence that olfactory function is severely damaged (Doty, 1997). A classic symptom of olfactory dysfunction are olfactory hallucinations which patients have described as "smells of holiness" or "poisonous vapours" (Leopold, 1995). In a study by Bromberg and Schilder (1934), it was found that the highest number of patients with an olfactory disorder were also sufferers of schizophrenia and alcohol psychosis (cited from Leopold, 1995). Although it is not thought to be sex linked, impaired olfaction is higher in males (Doty, 1997; Li et al., 1995). Testing olfactory function in schizophrenics can be used to assess brain function because their ability to identify odours is reduced.

 From the list of aetiologies described, it is clear that olfactory disorders can result from a wide range causes.

Treatments

Currently, there is no single therapeutic approach for all aetiologies, so instead, when looking at treatments we must look at each one independently. According to Jafek et al. (2000), "the therapy of olfactory loss is cause specific" therefore each aetiology will be considered separately followed by a discussion of the efficacy of each method of treatment.

 i) Nasal and/or Sinus Disease (NSD)

This aetiology is thought to be the most amenable to therapeutic interventions. The general consensus of opinion suggests that anosmic patients with allergic rhinitis and/ or nasal polyposis can be successfully treated following administration of systemic corticosteroids and intranasal topical corticosteroids (Cullen and Leopold, 1999; Smith and Seiden, 1991; Deems et al., 1991). If neither is thought appropriate for a patient, then surgery may be a carefully considered alternative.

a) Systemic corticosteroids

Systemic corticosteroids are potent anti-inflammatory substances that act by reducing the nasal mucous membrane, allowing the odorant to reach the olfactory neuroepithelium (Ikeda et al., 1994). Prednisonet is extremely effective in treating smell disorders and it is often used as a diagnostic tool. When writing about his own accounts with anosmia, Dr Karl Wuensch describes the therapy that he has experienced. The cause of his anosmia was unclear although he did suffer from nasal polyps and so was prescribed a 10mg dose of prednisone to be taken daily. Initially he experienced one episode of recovery and this gradually increased to a full recovery. Subsequently this dosage was reduced until, eventually he stopped taking the medication and his anosmia returned. Although this treatment did temporarily correct his disorder it was not a permanent solution in his case.

b) Intranasal topical corticosteroids

Adverse side effects associated with systemic corticosteroids include a raised blood pressure, therefore it is advisable to only take a short course of medication (Cullen and Leopold, 1999; Mott and Leopold, 1991; Smith and Seiden, 1991). Intranasal topical corticosteroids are a reasonable alternative. In the US these include beclomethasone dipropionate, flunisolide and dexamethasone sodium phosphate. They are believed to be more potent than systemic administrations, which rapidly metabolise into inactive or much less active metabolites. Local side effects are usually mild, for example mucosal dryness and sneezing (Whittet et al., 1991).

Golding-Wood et al., (1996) carried out a study involving fifteen patients who suffered from hyposmia with perennial rhinitis (often a neglected symptom). Each patient was asked to take three drops of betamethasone sodium phosphate daily, for six weeks and were evaluated using a UPSIT test before and after treatment. The results showed that all of the patients improved their test scores after treatment and therefore suggesting that topical steroid therapy is affective for treating seasonal perennial rhinitis and nasal polyps.

Nasal sprays have proven to be an effective method of treatment. The distribution of the spray is enhanced when taken in the 'head down forward' (HDF) or 'Moffat's position. It is believed to encourage maximum exposure of the drug to the nasal and paranasal sinus mucosa. However, why it is not effective to all, remains unclear (Chalton et al., 1985).

Mott et al. (1997), invited anosmic patients to complete a questionnaire about their olfactory complaint and then were asked to spray two doses of flunisolide into each nostril twice a day (using the HDF position) and a course of antibiotics to fight any bacterial infection. The results obtained 8-26 weeks after treatment, showed that within the group two-thirds reported olfactory improvements. Patients who experienced fluctuations in their smell sensitivity, had higher UPSIT scores than those who did not report such a finding. When describing why their treatment had been successful, Mott et al. (1997) highlighted areas such as patient compliance, the combination of antibiotics and the effective HDF position.

There are several factors that contribute to difficulties within NSD treatment, for example the high recurrence rate of nasal polyps and the inaccessibility of some of these processes to local therapy (Mott and Leopold, 1991). Topical steroid results may be limited because of substantial mucosal swelling in the upper nasal cavity which may prevent the delivery of the steroid (Golding-Wood et al., 1996). Symptoms often persist and there is approximately an 80% chance of recurrence rate of polyps, therefore olfactory loss can continue regardless of the therapy (Mott and Leopold, 1991).  

Alternative therapies such as herbal remedies, have been proposed to improve smell sensitivity. In one case an anosmic was given several natural supplements with the aim to rectify levels of trace elements including magnesium and zinc. Within five weeks the patient reported an improvement which continued until a complete recovery was observed. However, whether this evidence can be credited is debatable.  

c) Surgical Interventions

Surgical procedures have proven to reduce nasal obstruction and improve olfaction ability. Procedures including endoscopic ethmoidectomy can restore symptoms and in some cases these can last for up to two years (Smith and Seiden, 1991). It is usually considered to be the third therapeutic option because it can be very painful and there is no guarantee that it will be successful.  

ii) Treatments for olfactory loss from Head trauma(HT)/ Post traumatic injuries

As mentioned previously, the olfactory system has the ability to regenerate. Indeed, there is the potential for recovery after a head injury and animal studies have demonstrated that recovery is possible. Costanzo (1985, cited from Smith and Seiden, 1991), recorded cells from the olfactory bulb of a hamster and demonstrated recovery within nine months. However, in humans the prognosis is much lower and the estimated possibilities for recovery vary widely from 15% to 39%. (Duncan and Smith, 1995; Smith and Seiden, 1991).  

It is generally believed that this aetiology can not be treated with drug therapy. However, approximately a third of sufferers do recover (Duncan and Seiden, 1995) with the most likely cause being natural regeneration of the olfactory system. The onset of regeneration usually occurs within three months of the trauma (Mott and Leopold, 1991); beyond one year then the chances of recovery are slim (Ikeda et al., 1995). Early recovery may occur due to mechanisms such as the disappearance of blood clots, and later recovery, due to regeneration of neural elements (Smith and Seiden, 1991; Smith and Seiden, 1991). Complete recovery may take approximately five years (Duncan and Seiden, 1995).  

To test the hypothesis that medications may relieve symptoms, Ikeda et al (1995) invited seventeen patients with HT to take either a topical nasal drop of 0.1% betamethasone (A) or a systemic oral dose of prednisone (B). Three out of five from group B and only one from group A showed signs of recovery, therefore suggesting that topical corticosteroids are not significantly effective for treating head trauma. 

Levy et al. (1998) have recently proposed the use of theophylline, a nonselective phosphodiesterase (PDE) inhibitor (Beavo, 1995), as a therapeutic approach for relieving symptoms. Four patients with hyposmia; one HT and three NSD, were treated for four to six months. Theophylline acts by blocking the PDE involved in the transduction process, therefore preventing cAMP metabolising into AMP. Whether it can be considered as an effective therapeutic agent is controversial because toxic side effects have been reported at doses which are very close to therapeutic ranges (Beavo, 1995). However, theophylline has been used in asthma treatment for several years. 

Three patients did respond to treatment whilst one did not. Before and after treatment the patients were exposed to a fMRI scan which, in the presence of olfactory stimuli, highlighted the brain regions which are impaired in these patients.

Duncan and Seiden (1995) performed a long-term follow up study which looked at the likelihood of recovery in forty-one patients either HT or URI symptoms. Over a period of five years signs of recovery were seen in both groups therefore providing these patients with an alternative outlook. It has been proposed that patients with dysosmia are more likely to recover, but this was not the case in this study because although the dysosmics did have a higher UPSIT score, the amount of improvement was no greater than in non-dysosmics.  

During recovery, some anosmics have reported episodes of phantosmia or parosmia. Wright (1987) proposed that the occurrence of parosmia results from minimally functioning olfactory system (cited from Smith and Seiden, 1991). This may explain why HT patients reporting parosmia or phantosmia have higher UPSIT test scores then patients without these symptoms (Smith and Seiden, 1991). Currently the causes of phantosmia are not fully understood and treatments are still experimental (Leopold, 1995).  

Recently two approaches have been developed, drug therapy involving haloperidol and surgical interventions. As a neuroleptic, haloperidol controls hallucinations, thus inhibiting the phantom smells (Henkin et al., 2000). With regard to surgical interventions, excision of the olfactory mucosa has successfully been performed by Leopold et al. (1991) on a patient with unilateral phantosmia. Immediately after the operation the patient report no olfaction ability but then her symptoms gradually improved. They concluded from this two reasons why it had been a success; firstly, because the neurones that generated the unpleasant odours had been removed and secondly, that the olfactory epithelium excised had been connected to the olfactory bulb that incorrectly interpreted the incoming signals. Histological observations also showed that part of the epithelium was disconnected. They therefore be concluded that interrupting the incoming signals removed phantosmia.  

Unexpectedly, this procedure also fully restored the patient's sense of smell, possibly because normally functioning receptor cells were spared. However, these results come from a single case and further suport is necessary. For patients with severe dysosmia, Cullen and Leopold (1999) have performed olfactory cleft ablations. By completely removing the olfactory neurone cell bodies the symptoms can be eliminated. So far this procedure has produced some encouraging results. 

iii)Prior Upper Respiratory Infection (URI)

Currently, there is no effective medical treatment to restore the olfaction ability in these patients (Hummel, 2000; Archer, 2000; Duncan and Smith, 1995; Mott and Leopold, 1991). Instead, if recovery does occur then it is likely to be spontaneous. Evidence of spontaneous recovery is conflicting. Deems et al (1991) reported that patients with URI do not recover their olfactory ability. There is no hard evidence to support spontaneous recovery, nor is there any to suggest that it can not happen (Smith and Seiden, 1991), therefore highlighting that the mechanism and prognoses of the disorder is not fully understood. 

Approximately a third of patients do recover and in the majority of these cases, recovery occurs six months after onset (Hummel, 2000). However, because some patients do not seek medical advice after a cold, many sufferers may go unnoticed and hence the actual recovery rate could be around 50% (Hummel, 2000). The length of recovery time can be 3 years (Cullen and Leopold, 1999), however, this can only be speculation because animal models have shown that recovery time varies amongst species (Mott and Leopold, 1991).  

An alternative and some what controversial treatment involves the administration of natural trace elements such as zinc and vitamin A. Evidence involving zinc is rather conflicting, hence the term 'zinc controversy' (Duncan, 1995). Although it is quite common for patients with taste or smell disorders to be given zinc supplements, most patients do not notice any improvements in there symptoms (Deems et al, 1991). In a randomised, double blind zinc test Henkin et al (1976) compared the effects of zinc sulphate with placebo in patients with smell and taste disorders. The results showed no therapeutic affect in the olfactory patients. 

Treatment involving vitamin A is not generally supported. In a study performed by Duncan and Briggs (1962), patients with uncomplicated anosmia (i.e. excluding atrophic rhinitis) were treated with a dose of vitamin A and the majority responded well. From this evidence, it was suggested that vitamin A acted by regenerating the olfactory cells of the mucous and serous glands.  

Conclusion

Why do some patients regain their olfactory function? Is it that once a particular level of disruption is reached, no recovery will be seen? It is a commonly held view that the greater the damage to the neuroepithelium, the lower the chances of recovery, but is this really the case? Although it can not be claimed that complete recovery is inevitable, further studies investigating the mechanisms of damage may bring new light to this opinion. With regards to this, an interesting hypothesis has been put forward that suggests that if cells do degenerate and then regenerate, then either one of these processes is slower than first thought or more complete regeneration is needed to regain the ability of smell (Duncan and Seiden, 1995). 

An important area to be considered is time-scale. Retesting after one year is not long enough (Duncan and Seiden, 1995). Currently there is no standard length of time amongst the surveys, some retest patients after a few weeks whilst others (long-term follow up) studies reassess after several years. These longer studies must be more accurate and beneficial because they provide more information about recovery mechanisms which is area of on going research. Knowing whether recovery can and does occur may help us to understand the mechanism of olfactory loss in these patients (Duncan and Seiden, 1995). Future research should involve retrospective studies where questionnaires are sent out to patients post therapy because some important information with regard to recovery may be lost otherwise.  

One reason for the conflicting opinions depends on the methodical approach used. For example, the studies before the introduction of the UPSIT test, may have missed some information that the presence of UPSIT may have found. Using neuroradiologic equipment is the diagnostic tool of the 21st century and should be used to diagnose more (all) patients (if financially viable.

All patients should be entitled to counselling and be made aware of adjustments for everyday living, for example the installation of smoke detectors. 

It should be remembered that not all patients respond to the same treatments. Jafek et al, (1987) reported a case where a patient responded when given a combination of corticosteroids and surgery but not when they are administered independently. This emphasises the difficulties of patient prognosis. 

When recovery does look unlikely, then it is important that patients develop adaptive strategies so that they can cope with personal hygiene, appetite, safety and health. Other areas that should also be considered are vocational, psychological and cognitive difficulties because to some patients they can lead to great anxiety.

 More people should be educated about the symptoms and causes of anosmia so that smell no longer remains the 'forgotten sense'. But, as with so many disorders, if you do not think that it is going to affect you personally, are you really that concerned?

REFERENCES

ARCHER, S.M. (2000). The evaluation and management of olfactory disorder following upper respiratory tract infection. Arch. Otolaryngol Head Neck Surgery 126, 800-801. 

BEAVO, J.A. (1995). Cyclic nucleotide phosphodiesterases: functional implications of multiple isoforms. Physiol. Rev 75, 725-748. 

CULLEN, M.M. & LEOPOLD, D.A. (1999). Disorders of smell and taste. Med. Clin. of N. America 83, 57-74. 

DAVIDSON, T.M., MURPHY, C. & JALOWAYSKI, A.A. (1995). Smell impairment can it be reversed? Postgraduate Medicine 98, 107-118. 

DEEMS, D.A., DOTY, R.L., SETTLE, G., MOORE-GILLON, V., SHAMAN, P., MESTER, A.F., KIMMELMAN, C.P., BRIGHTMAN, V.J. & SNOW, J.B. (1991). Smell and taste disorders, a study of 750 patients from the university of pennsylvania smell and taste center. Arch Otolaryngol Head Neck Surgery 117, 519-528. 

DOTY, R.L. (1997). Studies of human olfaction from the university of pennsylvania smell and taste center. Chemical Senses 22, 565-586. 

DUNCAN, H.J. & SEIDEN, A.M. (1995). Long-term follow-up of olfactory loss secondary to head trauma and upper respiratory tract infection. Arch Otolaryngol Head Neck Surgery 121, 1183-1187. 

DUNCAN, H.J. & SMITH, D.V. (1995). Clinical Disorders of Olfaction: A review. In DOTY, R.L. Handbook of olfaction and gustation. Marcel Dekker Inc. New York, USA. pp 345-366. 

FELDMAN, J.I., WRIGHT, H.N. & LEOPOLD, D.A. (1986). The initial evaluation of dysosmia. Am. J. of Otolaryn. 4, 431-444. 

GOLDING-WOOD, D.G., HOLMSTROM, M., DARBY, Y., SCADDING, G.K. & LUND, V.J. (1996). The treatment of hyposmia with intranasal steroids. Journal of Laryngo. & Oto. 110, 132-135. 

HENKIN, R.I., LEVY, L.M. & LIN, C.S. (2000). Taste and smell phantoms revealed by brain functional MRI (fMRI). J. of Computer Ass. Tomography 24(1), 106-123. 

HENKIN, R.I., MARTIN B.M. & AGARWAL R.P. (1999). Efficacy of exogenous oral zinc in treatement of patients. Am. J. Medical Sciences 318, 392-305. 

HENKIN, R.I., SCHECTER, P.J., FRIEDEWALD, W.T., DEMETS, D.L. & RAFF, M. (1976). A double blind study of the effects of zinc sulfate on taste and smell dysfunction. Am. J. Medical Sciences 272(3), 285-299. 

HUMMEL, T. (2000). Perspectives in olfactory loss following viral infections of the upper respiratory tract. Arch. Otolaryngol Head Neck Surgery 126, 802-803. 

IKEDA, K., SAKURADA, T., TAKASAKA, T., OKITSU, O. & YOSHIDA, S. (1995). Anosmia following head trauma: Preliminary study of steroid treatment. Tohoku Journal of Experimental Medicine 177, 343-351. 

IKEDA, K., SAKURADA, T., SUZAKI, Y. & TAKASAKA, T. (1994). Efficacy of systemic corticosteroid treatment for anosmia with nasal and paranasal sinus disease. Rhinology 33, 162-165. 

JAFEK, B.W., MORAN, D.T., ELLER, P.M., ROWLEY ILL, J.C. & JAFEK, T.B. (1987). Steroid-dependent anosmia. Arch Otolaryngol Head Neck Surgery 113, 547-549. 

JAFEK, B.W., MURROW, B.& LINSCHOTEN, M. (2000). Evaluation and treatment of anosmia. Current Opinion Otolaryngo & Head & Neck Surgery 8, 63-67. 

LEVY, L.M., HENKIN, R.I., LIN, C.S., HUTTER, A. & SCHELLINGER, D. (1998). Increased brain activation in response to odours in patients with hyposmia after theophylline treatment demonstrated by fMRI. J. of Computer Ass. Tomography 22(5), 760-770. 

LI, C., YOUSEM, D.M., DOTY, R.L. & KENNEDY, D.W. (1995). Evaluation of olfactory deficits by medical imaging. In DOTY, R.L. Handbook of olfaction and gustation. Marcel Dekker Inc. New York. USA. pp395-419. 

MOTT, A.E., CAIN, W.S., LaFRENIERE, D., LEONARD, G., GENT, J.F. & FRANK, M.E. (1997). Topical corticosteroid treatment of anosmia associated with nasal and sinus disease. Arch Otolaryngol Head Neck Surgery 123, 367-372. 

NICKELL, W.T. (1997). Basic anatomy and physiology of olfaction. SEIDEN, A.M. In Taste and smell disorders. Thieme. New York. USA. pp 20-37. 

LEOPOLD, D.A. (1995). Distorted Olfactory Perception. Ed. DOTY, R.L. In Handbook of Olfaction & Gustation. Marcel Dekker Inc. New York. USA. pp441-454.  

LEOPOLD, D.A. (2000). A perplexing olfactory loss. Arch. Otolaryngol Head Neck Surgery 126, 803. 

LEOPOLD, D.A., SCHWOB, J.E., YOUNGENTOB, S.L., HORNUNG, D.E., WRIGHT, H.N. & MOZELL, M.M. (1991). Successful treatment of phantosmia with preservation of olfaction. Arch Otolaryngol Head Neck Surgery 117, 1402-1406. 

SEIDEN, A.M. (1997). Taste and Smell Disorders. Thieme. New York. USA.  

SMITH, D. V.& SEIDEN, A.M. (1991). Olfaction dysfunction. Ed. LAING, D.G., DOTY, R.L. & BREIPOHL. In The Human Sense of Smell. , W. Sprinzer-Verlag, Berlin. pp283-305. 

THOMPSON, M.D., KNEE, K. & GOLDEN, C.J. (1998). Olfaction in persons with alzheimer's disease. Neuropsychology review 8, 11-23. 

VAN TOLLER, S. (1991) Assessing the impact of anosmia: review of a questionnaire's findings. Chemical Senses 24, 705-712.

WHITTET, H.B., SHINKWIN, C. & FREELAND, A.P. (1991) Anosmia due to nasal administration of corticosteroid. Letters BMJ 303, 651.