Australasian Science Magazine

Smell and taste disorders compromise the nutritional health of a significant proportion of young Australians.

By David Laing

The rate of taste disorders in children exceeds World Health Organisation guidelines, and combined with smell disorders compromises the nutritional health of a significant proportion of young Australians.

If you have a poor sense of smell you will not detect the presence of leaking gas in the home or service station, or the smell of other poisonous gases like chlorine and the foul-smelling rotten egg gas, which can be emitted at dangerous levels from swamp lands or marshes.

Poor sensitivity also means that you will perceive the flavour of foods differently to other people. For example, if you cannot detect orange, apple or strawberry smells, these fruits will have little flavour and be bland. Similarly, if you cannot perceive the basic tastes of sweet, salty, sour, bitter and umami – the savoury taste of monosodium glutamate in many Asian dishes – food will be perceived as bland and unpleasant.

Changes to the flavour of food and beverages as a result of smell or taste loss can have significant effects on health and result in inadequate nutrition, anorexia or obesity. These serious outcomes arise because people find it difficult to adjust to their new world of unpleasant foods. Some people, however, continue to eat normally, and there is some evidence that these individuals can imagine flavours.

Loss of smell or taste can be due to diseases, head injuries, medications, or people can be born smell-blind or taste-blind. Often loss of these senses is temporary and may last up to a year or two.

Fortunately, unlike the senses of vision and hearing, smell and taste receptor cells in the nose and tongue are continuously regenerating. All of the taste cells in the tongue, for example, die and new ones are produced about every 2 weeks, while all of the smell cells in the nose are replaced within a few weeks or up to a year depending on the environment in which a person lives. Chemical, dusty or smoky work environments will result in shorter replacement periods.

Smell Disorders in Children
Diagnosing smell or taste disorders is difficult for clinicians since there has been no universal test they can use to assess patients. This means there is little information about how many people in a community have smell or taste disorders.

Although tests of smell have been developed in the USA and Germany over the past decade for use with adults, they are not suitable for assessing children. Accordingly, during the past decade one of the main aims of our research has been to resolve this situation and develop tests to establish the prevalence of smell and taste disorders in children.

Our goal was to have tests that were sensitive, simple, rapid and portable for use at the bedside in hospitals and outpatient departments and in large-scale screening studies that could be conducted in schools. The focus has been mainly on children of school-age. Since a test involving the identification of smells provides the most reliable procedure, we set out to identify which smells were well-known to Australian children.

To achieve this we had hundreds of children and university students smell several hundred foods and beverages, fruits and vegetables, and commonly encountered smells from household products such as soaps and common cleaning agents, petrol, paint, burnt products with smoky smells, and the aromas of perfumes and flowers. From these we pinpointed about 40 smells that were readily identified by the majority of children and students.

To improve the identification rate, we also asked them to choose one word from a set of four words that best represented the smell. We expected this would improve the accuracy of identification because of the well-documented “tip of the nose” effect, which is derived from the common expression: “I know the smell but can’t put a word to it”.

Finally, we produced a test that uses 16 common food and non-food smells in which a person chooses one labelled photograph from a set of three that best represents the smell. Sixteen sets of three photographs were included in a booklet that was used to identify smells sniffed from squeeze-bottles.

Data obtained from hundreds of healthy children indicate that 5-, 6- and 7-year-olds with a normal sense of smell will identify at least 11, 12 and 13 of the 16 smells, respectively, and that there is a prevalence of 1.9% of smell disorders in the general population of children in Australia.

The Taste Test
The development of a taste test for children aged 5 years and above followed a similar pathway, but because there are only four commonly recognised tastes compared with thousands of smells, this was achieved more rapidly. In the taste test, a person sips a few millilitres of water, as well as sweet, salty, sour or bitter water solutions, and identifies the taste from a set of three labelled photographs in a booklet. Each taste quality (e.g. sweet) is prepared in five concentrations that range from weak but identifiable to very strong, providing a total of 25 solutions (five solutions are of water alone).

Our studies show that people have a taste disorder if they incorrectly identify three of the five concentrations of a taste. For example, if someone can’t taste sweetness at moderate strength then sweetness would not be perceived in many foods, including most vegetables and some fruits.

Failure to identify four of the concentrations of a taste indicates that the taste will only be perceived in the sweetest foods, such as some confectionaries. People are classified as taste-blind to a quality if they are unable to identify the strongest solution of a taste.

Since 60–70% of foods and beverages contain some sweetness, people with a sweet disorder will find most fruits sour or bitter and many foods unpleasant.

The smell and taste tests have been used in a range of studies of more than 2000 children in schools and at Sydney Children’s Hospital. At the hospital we found that about one-third of children with chronic renal disease have taste disorders and that there is a strong correlation between the extent of kidney failure and taste loss. This outcome appears to account for the reluctance of affected children to eat and maintain adequate nutrition.

Surprisingly, we found that, contrary to anecdotal data, children with cystic fibrosis had normal smell and taste function until at least 18 years of age. It had been thought that the thick mucus that typically accumulates in the nasal cavity and respiratory system of children with cystic fibrosis would block access to smell receptors in the nose.

Similarly, we have shown that children with various forms of cancer or who have had chemotherapy or radiotherapy do not have smell or taste disorders, ruling out these senses as the cause of their reluctance to eat over extended periods of time.

Sensory Disorders in Aboriginal Children
Recently we conducted the first ever studies of smell and taste function in Aboriginal children and compared their responses with non-Aboriginal children. Our interest in investigating the two senses in Aboriginal children arose from the very high incidence of middle ear disease (otitis media), which affects up to 90% of Aboriginal children. Otitis media also has a high prevalence in non-Aboriginal children (30%) and is the most common reason for children in the Western world to visit a general practitioner (18% of all visits). Otitis media also causes loss of hearing in many children.

The reason for our interest was twofold. First, one of the major taste nerves in the tongue passes through the middle ear en route to the brainstem, where the first processing of taste information occurs. Importantly, histopathological studies have shown that this taste nerve is attacked by the bacteria and viruses that are present in the middle ear during otitis media, and there have been unconfirmed reports that the resulting damage alters or destroys taste function. Accordingly, our goal was to determine whether Aboriginal children exhibited high levels of taste loss.

Secondly, we assessed smell function because the bacteria and viruses in otitis media are found in the upper respiratory tract where smell receptor cells reside, and chronic rhinitis and sinusitis brought about by these pathogens can cause smell disorders.

Our initial study with 300 Aboriginal children aged 9–12 years from 16 communities across NSW indicated that only one child could not smell at all and only four others had slightly poorer smell sensitivity than normal.

In contrast, 8% of children had taste disorders characterised by an inability to perceive one or more of the basic taste qualities, and 20% had hearing loss. This latter finding provided strong evidence that many of these children had experienced otitis media.

However, the taste test used was a simpler and less reliable than the test described above. Accordingly, to confirm the high prevalence of taste disorders we conducted a second study using our more thorough taste test. In this instance we assessed taste function in 432 Aboriginal and non-Aboriginal children aged 8–12 years from six public schools in Dubbo. The children were matched for age, gender, education and general living environment.

The results indicated that significantly more Aboriginal children (12%) had taste disorders than non-Aboriginal children (8%), suggesting that their more frequent exposure to otitis media was the likely cause of the difference. Interestingly, there was no difference in the prevalence of taste disorders in females and males.

An important implication of these results, however, is that the level of disorders in both groups of children exceeds the level set by the World Health Organisation (4%) for triggering action by government health authorities to investigate and reduce the prevalence of a disorder or disease in a population. Given the implications of taste disorders for the nutrition and general health of the Australian population there is a clear need for a government program to address the causes, consequences and management of taste disorders in children to minimise their effects later in life.

Prevalence of Sweet Disorders
In addition to the finding of a high level of taste disorders in Australian children, the results may have led to a serendipitous finding regarding how humans code taste qualities and that the sweet quality is the dominant one in humans. Of the 41 (9.5%) children with disorders in the Dubbo study, 27 (65.9%), 17 (41.5%), eight (19.5%) and five (12.2%) had sweet, bitter, salt or sour disorders, respectively. Furthermore, in every instance of multiple disorders the sweet quality was involved.

The major question arising from these results is why were sweet disorders the most prevalent type, and why was there an unequal representation of the four taste qualities?

The unequal numbers could have occurred if the neural processing systems for each taste quality differ in ways that make a specific quality more vulnerable than the others to particular diseases, trauma or medications. As far as is known, the nerves that carry information about the sweet quality are not more vulnerable to otitis media or any other disease than the nerves for the other qualities.

Another possibility is that different taste qualities have different numbers of taste receptor cells and nerve fibres from the tongue, so if sweet tastes have more cells there would be a greater chance of these being affected by diseases. Alternatively, if there are equal numbers of taste cells for each taste quality we should have found equal numbers of disorders across the four qualities. Clearly this did not occur.

To resolve this question we looked at taste in different species. Phylogenetic studies indicated that there are wide differences in taste between species, and that the closer a species is related to humans the greater the number of sweet cells and nerves occur. Indeed, approximately 50% of taste nerve fibres in the chimpanzee, the closest relative to humans, are sweet-sensitive. Furthermore, humans and chimpanzees respond similarly to a variety of natural and artificial sweeteners. In contrast, not all sweeteners taste sweet to the cat, hamster, pig and rhesus monkey.

The chimpanzee, therefore, is the only species in which studies of its sense of taste have not yet revealed any substantial differences from that of humans. Thus, we propose that humans may have a similar representation of the four taste qualities as the chimpanzee, with sweet the dominant quality. We also propose that, given the large number of children in the study and the possible variety of diseases, traumas and medications that could have produced disorders during their lifetime, the number of disorders recorded for each taste quality reflect the representation of nerve fibres for each quality in the human taste system.

These proposals have yet to be proven, but they provide two unique explanations regarding the structure and functioning of the human sense of taste.