Fluoride in toothpaste - is the expressed total fluoride content meaningful for caries prevention?

• This article gives an overview of the evolution of fluoride toothpaste formulations, which have become increasingly complex, and those produced in developing markets may be of both variable quality and storage properties.

• The total fluoride content declared by manufacturers does not necessarily translate into bioavailable ionic fluoride on brushing.

• If manufacturers declared the free ionic fluoride in parts per million available on brushing, rather than the total theoretical content of the formulation, it would provide a more meaningful indicator of efficacy that dental professionals, pharmacies and consumers could rely upon.

The aim of this paper is not to provide a review of the efficacy of fluoride toothpastes for improving oral health; nor is it a commentary on the claims of tooth whitening, plaque control or tartar prevention made by manufacturers for an ever-changing and increasing range of products available on the global market. The aim is firstly to give a brief overview of the evolution of fluoride toothpaste formulations and then to address two questions: is the total fluoride content of a toothpaste the most meaningful for caries prevention, and if not, how can the fluoride content be most reliably and meaningfully expressed for the guidance of health professionals and consumers?

In 1945, the renowned paediatric dentist and researcher Basil Bibby suggested that a fluoride-containing toothpaste might be an effective product for caries prevention.¹ Following research during the 1930s and 1940s showing that fluoride present in drinking water supplies was associated with the inhibition of caries, the potential advantage of topical application in the form of a fluoride-containing toothpaste was put forward. Early experimental work using calcium carbonate (chalk) as the abrasive base proved ineffective, as the fluoride reacted with the base and precipitated as insoluble calcium fluoride. Following work by Joseph Muhler and William Nebergall at the University of Indiana,² the first commercially successful fluoride-containing toothpaste approved by the American Dental Association (ADA) was launched in 1954 by the Procter & Gamble Company under the brand name 'Crest with Fluoristan'. This product contained stannous fluoride - tin(II) fluoride - as the active ingredient. The fluoride salt was later changed to sodium monofluorophosphate (MFP), possibly because of the rising cost of tin.

Over the intervening years, it became accepted that fluoride-containing toothpastes were by far the most important way of delivering the beneficial effect of fluoride.³ The marketing of toothpaste, and indeed other oral health products, has become highly competitive and an important source of revenue for the major international manufacturers. It should be remembered that, in the UK, the very first product to be advertised on commercial broadcast television 12 minutes after it began at 8 pm on 22 September 1955 was a toothpaste. In today's global market, there is a very wide range of fluoride toothpastes varying in formulation, fluoride concentration, quality control of ingredients and manufacture, and some products are transported, stored and sold to consumers in the countries far removed from where they are made.

Today's toothpastes in developed markets are very complex products with a multitude of formulations. All are aqueous suspensions of abrasive compounds and include active ingredients, flavour compounds, artificial sweeteners, colourings, binding gums, preservatives and other excipients.⁴ The active ingredients include anti-caries agents (fluorides), anti-malodour agents, anti-tartar (calculus) agents, anti-plaque/anti-gingivitis agents, whitening agents, agents for the relief of dentine hypersensitivity and erosion prevention agents. The competitiveness of the toothpaste market has driven the major manufacturers to offer an increasing and frequently changing variety of formulations to support claims of caries prevention, tooth whitening, sensitivity prevention, gum health and total oral health. However, the focus of this article is to consider how variations in formulation and the fluoride content of toothpastes might affect clinical efficacy for caries prevention.

A recently published review of clinical trials in Cochrane Database of Systematic Reviews showed only weak evidence of increasing benefit with total fluoride concentrations in the reviewed studies.⁵ This may be due to a wide range of factors. Methodological variations in diagnostic criteria, population characteristics and sampling methods are important factors. However, variations between different formulations having the same total fluoride concentration can have an impact on caries reduction. It must be remembered that, in some clinical trials, specially produced standardised toothpaste formulations differing only by fluoride concentration have been compared, while other studies have compared different commercially available toothpastes with differing fluoride concentrations. As toothpaste formulations evolve and become more complex, partly driven by marketing requirements, we must now question whether the total fluoride content as expressed in manufacturers' descriptions are fully and directly related to caries reduction.

Following their introduction, fluoride toothpastes have been marketed with a range of total fluoride content from under 500 ppm to 5,000 ppm. The effectiveness of toothpastes with less than 1,000 ppm fluoride is uncertain due to the limited number of studies available. However, several studies have indicated that low fluoride formulations introduced for use by children were not as effective for caries prevention in primary teeth as those in the range of 1,000-1,500 ppm.^5,6 More research into the efficacy of low-fluoride toothpastes is needed, particularly on primary teeth. However, a further clinical trial testing 500 ppm vs 0 ppm is very unlikely to happen due to strong ethical reasons and the best conclusions for public health must come from the available evidence. The current advice in the UK, but not all countries, is that toothpastes with a fluoride content of at least 1,000 ppm should be used by both children and adults, although just a smear of paste for infants and a pea-sized amount for children.^7,8,9

Of the three inorganic fluoride compounds commonly used in toothpastes, sodium fluoride is the most soluble (40 g/100 ml), dissolving rapidly in water to give sodium fluoride ions (F^-)^. Stannous fluoride is somewhat less soluble (35 g/100 ml), while MFP is the least soluble (25 g/100 ml), hydrolysing slowly in vitro to liberate fluoride ions. This quality makes MFP more compactible with a calcium carbonate abrasive base, producing a relatively inexpensive formulation. Over the years, there has been a trend to use sodium fluoride as the active ingredient, sometimes in combination with MFP or stannous fluoride, with inert silica as the abrasive base. While the range of commercially available toothpastes worldwide use all three inorganic fluoride compounds, and to a lesser degree the organic amine fluorides, sometimes in combination, the most commonly used base remains silica, which is compatible with all three inorganic fluoride compounds. Some formulations include dicalcium phosphate dihydrate or titanium oxide, often in combination with other abrasive bases. Added to this, there may be variations in the quality grades of the ingredients used in different manufacturing facilities globally, some of which may contain trace impurities that can affect fluoride ion concentration, and some formulations include anti-nucleation agents such as pyrophosphates, aimed at reducing the formation of calculus (tartar). Globalisation has produced a complex market for toothpaste where familiar brands can be manufactured thousands of miles away from their traditional customer base, and familiar packaging can hide a change in both formulation and possibly efficacy.

Research over the past 60 years has shown that fluoride produces its effect in a number of different ways which combine to slow and help prevent the caries process and also to reverse caries in its early stages.¹⁰ These are given below in the order of effectiveness:

• Enhanced remineralisation - in vitro studies by McCann and Brudevold working in Boston in the 1960s showed that fluoride ion concentrations as low as 0.2 ppm favoured the precipitation of hydroxyapatite from calcium phosphate solution and the growth of existing hydroxyapatite crystals.¹¹ This work lead to the concept that low levels of fluoride in the plaque and saliva are able to alter the chemical balance between demineralisation of the enamel and remineralisation.¹² The effect favours the remineralisation process, allowing the early carious attack on enamel to be reversed, and new mineral crystals with better structure and greater acid resistance to be deposited. This is the mechanism by which fluoride toothpaste is thought to work and appears to be the most important

• Reduced acid production - fluoride is concentrated in the plaque layer on the surfaces of the teeth and reduces the enzymatic conversion of dietary sugars into acid by plaque bacteria. Fluoride toothpaste also invokes this mechanism

• Fluoride substitution - fluoride entering the developing teeth from the diet via the bloodstream is incorporated into the forming hydroxyapatite crystals in enamel and dentine. The partly fluoridated hydroxyapatite that is deposited is theoretically more resistant to acid attack than that crystallised without fluoride

• Reduced pit and fissure depth - the parts of the teeth most susceptible to caries are the natural pits and grooves, or fissures, on the biting surfaces of back teeth. Fluoride entering the developing teeth at an early stage appears to result in reduced pit and fissure depth.

The use of fluoride toothpaste, which delivers its effect at the tooth surface, has reduced the significance of the last two mechanisms, which are now thought to play a minor role.

In the mouth, there is a constant and fluctuating exchange of mineral ions between the tooth surfaces and the covering plaque, a mechanism described as an 'ionic see-saw'. In a caries-free mouth, there is an equilibrium of ionic exchange.¹² In simple chemical terms, caries can be considered as resulting from an imbalance between periods of demineralisation of the enamel surface, when plaque pH is depressed by bacterial metabolism of dietary sugars, and subsequent remineralisation, when plaque pH returns to neutrality. Caries occurs when, over a period of time, frequent demineralisation/remineralisation cycles result in a loss of the ionic equilibrium between the enamel surface and the plaque and saliva, resulting in a net demineralisation of the enamel surface and cavity formation. As indicated above, the availability of low concentrations of fluoride ions in the oral fluids will not only help to inhibit plaque acid production and the demineralisation process, but will also favour remineralisation, thereby tilting the ionic exchange of mineral irons in favour of the enamel surface.

Furthermore, there is evidence that the presence of fluoride ions in the plaque during demineralisation/remineralisation cycles will progressively improve the structure of the hydroxyapatite crystals in the enamel surface, increasing resistance to demineralisation.¹⁰ The regular use of fluoride toothpaste facilitates these beneficial effects, as does an optimum fluoride level in the drinking water supply. While considered to be less important, fluoride absorbed into the bloodstream from drinking water and dietary sources during tooth formation may invoke the latter two mechanisms mentioned above. It is inevitable that, after brushing with fluoride toothpaste, a small but variable amount of fluoride will be ingested by young children. Manufacturers rightly place guidance on packaging to help avoid excessive ingestion that might lead to fluorosis affecting the permanent dentition.

There are a number of expressions for the fluoride content of toothpastes that are sometimes quoted by manufacturers or from independent analyses:¹³

• Total measurable fluoride - by current methods of analysis

• Declared total fluoride content - the amount stated by the manufacturers, usually as a percentage by weight for each fluoride compound in the formulation and/or converted to parts per million of fluoride ion (ppm F^-)

• Potentially available fluoride - the proportion of total fluoride that is soluble in water or acid solution

• Potentially bioavailable fluoride - the proportion of total fluoride that is chemically insoluble but may be soluble in oral fluids and may be released and available during tooth brushing

• Unavailable fluoride - the proportion of total fluoride that is insoluble in water or acid solution, but might be absorbed onto oral surfaces and become bioavailable over a period of time

• Soluble fluoride - a fraction of the total fluoride that becomes fully ionic in oral fluids

• Available fluoride - the amount, not usually stated on packaging or in manufacturers' product information, that becomes bioavailable in the mouth under standardised conditions after a defined period of time. The test approved by the ADA gives this period as one minute.¹⁴

The concentration of bioavailable free fluoride ion (F^-) in the oral fluids is the main determinant of efficacy. The concentration of fluoride ion immediately present in the mouth during tooth brushing, and the concentration in the mouth for a period of time afterwards, are dependent on a number of factors apart from the free ionic fluoride in the toothpaste as manufactured. The amount of toothpaste placed on the brush and the duration of brushing are clearly relevant variables. However, rinsing the mouth immediately after brushing is likely to rapidly reduce the concentration in the oral fluids, a factor supported by several clinical studies.^15,16,17,18

The complexity of different commercially available toothpastes causes a complication in comparing the initial fluoride ion concentration in the mouth and over time following tooth brushing. Formulations containing only sodium fluoride with an inert silica base will deliver a high initial percentage of their total fluoride content as ionic fluoride to the tooth surface during active tooth brushing. However, this is not the case for a similar formulation containing MFP, which must be broken down by the process of hydrolysis, liberating fluoride ions more slowly. This may be the explanation for the results from a meta-analysis showing greater caries reduction with toothpastes containing sodium fluoride rather than MFP.¹⁹ While sodium fluoride is a relatively small molecule with a molecular mass of 42 and is capable of rapid diffusion into plaque, MFP with a molecular mass of 144 diffuses more slowly into plaque, where it is hydrolysed to liberate fluoride ions. However, as any MFP retained in plaque slowly hydrolyses to release fluoride ions, ten minutes after brushing, fluoride ion concentrations in the oral fluids could be similar to those from a sodium fluoride formulation.²⁰ Both computer modelling and in vitro studies of MFP hydrolysis suggest that diffusion of MFP through plaque on the enamel surface and subsequent enzymatic hydrolysis by the action of plaque bacteria are dependent on plaque thickness, plaque pH and the microbial species present within the plaque.^21,22

The use of a calcium-containing base, such as calcium carbonate, in a toothpaste formulation creates a further complication. With a carbonate-containing base, insoluble calcium fluoride is precipitated from both sodium fluoride and hydrolysed MFP. However, if the mouth is not rinsed immediately afterwards and fluids are not consumed for several hours, then any calcium fluoride retained in the oral cavity could provide a reservoir for the slow release of fluoride ions. For a formulation containing only MFP and a calcium carbonate base, only a modest concentration of fluoride ion would be present on tooth brushing, but these fluoride ions would then react with the base to precipitate insoluble calcium fluoride. Once again, in the absence of rinsing and fluid consumption, hydrolysis of the retained MFP may liberate more fluoride ions, as might dissolution of the precipitated calcium fluoride, providing a slow release of fluoride ions.

It follows from these examples that the total fluoride content, commonly quoted by manufacturers for a product, may not be the most meaningful expression, as the free ionic concentration generated during tooth brushing can vary according to the formulation and also over time following tooth brushing. Only a toothpaste containing a fluoride agent that is readily soluble, such as sodium fluoride combined with an inert base such as silica, is likely to provide an immediate fluoride ionic concentration close to the total fluoride content quoted by the manufacturer. Nevertheless, it is possible that a formulation that provides a slow release of fluoride ion, beyond the period of active tooth brushing, may be equally or indeed more effective for caries prevention. This effect will depend on the clearance time within the mouth, as determined by saliva flow rate and any rinsing or fluid consumption in the immediate post-brushing period. At the present time, the effectiveness for caries prevention of any toothpaste formulation that can potentially provide a slow release of ionic fluoride is speculative, as there is little clinical evidence in the published literature to support such a mechanism. Any evidence would have to come from extended clinical trials. As can be seen, the dynamics of fluoride ion release from toothpaste are complex and cannot be determined by either the type of fluoride compound or its concentration alone. Tenuta and Cury²³ give a comprehensive review of toothpaste testing methods for bioavailability of fluoride in plaque biofilm.

Toothpaste randomised controlled trials (RCTs) are fraught with difficulties. Some investigators and their commercial sponsors have conducted trials in developing countries where recruitment might be easier and exposure to fluoride from other sources might be minimised. However, a wide range of factors confound any comparison of the results from different studies. Methodological variations in diagnostic criteria, population characteristics and sampling methods are some of the most important factors. Manufacturers cannot be expected to undertake extended RCTs for every new formulation, or for that matter when production is transferred to other facilities where the quality control of ingredients and production standards may vary.

A toothpaste made in Poland under a major international brand and widely available in the UK market, with a declared fluoride content of 0.32% sodium fluoride in a silica base, is stated to provide 1,450 ppm of fluoride ion (F^-). As indicated above, this figure is likely to be approached on brushing because of the solubility of sodium fluoride and the inert silica base. A second product, also made in Poland under the same brand, has a declared fluoride content of 1.1% MFP in a calcium carbonate base and is stated to provide 1,450 ppm fluoride ion (F^-). While the theoretical fluoride content of the MFP molecule is 0.132% by weight, a product containing 1.1% MFP would yield 1,450 ppm of fluoride ion, but only if fully hydrolysed. However, as indicated above, the ionic fluoride liberated from this formulation immediately on brushing is unlikely to approach this figure, due to the slow hydrolysis of MFP and precipitation of calcium fluoride from reaction with the calcium carbonate base. Consequently, the free fluoride ion trajectory of this product in the minutes after brushing becomes difficult to predict in vivo and is dependent upon a range of factors, including the physical and microbiological characteristics of the plaque and the chemical composition and flow rate of saliva.

There is persuasive evidence that the free ionic fluoride consideration in toothpastes varies with storage time and conditions. A comprehensive study of 16 adult toothpastes from Germany found that, of the pastes containing MFP, the free fluoride ion concentration was less than 20% of the manufacturer's declaration on production. After four months of storage, more than 75% of the ionic fluoride found in the new products could still be detected and the pastes containing MFP showed an increase in ionic concentration.²⁴ This finding may be due in part to continued non-enzymatic hydrolysis of MFP on storage. A study from Brazil showed that, after nine months of storage of toothpaste in schools, the total soluble fluoride decreased by an average of 21.9%.²⁵ A study from Chile of seven different toothpastes found that, on storage at room temperature, the loss of soluble fluoride from the tested samples could reach 40%.²⁶

A comprehensive study of 116 different toothpastes from the Netherlands and five countries in Southeast Asia found that, with one exception, all samples from the Netherlands complied with International Organisation for Standardisation (ISO) labelling requirements, and there were no differences between the declared fluoride content and that found to be present on analysis. In samples purchased in the other countries, the majority of MFP toothpastes showed a lower percentage of free available fluoride and most toothpastes did not follow standard labelling guidelines.²⁷ In a study of samples of 12 toothpastes on the Indian market, four toothpastes were found to have lower total fluoride than declared by the manufacturer and two had higher total fluoride. Four samples had considerably lower total soluble fluoride than total fluoride concentration.²⁸ A second Indian study of five adult and five children's toothpastes found that the mean percentage of total soluble fluoride in MFP/calcium carbonate toothpastes was 86% of the declared concentration, while that in sodium fluoride/silica-based toothpastes was 98%.²⁹ Analysis of seven different toothpastes marketed in West Africa and containing MFP showed mostly lower free fluoride concentrations than declared, especially true of those containing calcium-based abrasives.³⁰ A larger study of 22 fluoride toothpastes marketed in South Africa showed that the total fluoride concentration in all of the toothpastes was lower than that declared by the manufacturers, with one in four having a total soluble concentration of less than 1,000 ppm F. The total fluoride concentration for the calcium-containing toothpastes was lower than for the silica-based products.³¹

In their 1997 overview of developments in fluoride toothpastes, Holt and Murray stated that manufacturers should: continue to improve the performance of fluoride toothpaste; ensure that all pastes maximise fluoride bioavailability; develop active agents to help reduce oral disease; label products clearly with ppm F; and review delivery systems so as to reduce the risk of dental fluorosis.³² Although these aims have partly been achieved by the majority of the major manufacturers supplying established markets, they have yet to be achieved in many countries where regulatory controls and manufacturing standards are still evolving.

On the basis of presently available evidence, it would appear that the most reliable indicator of the effective fluoride content for a toothpaste formulation is the ionic fluoride concentration (F^-) as parts per million available immediately on brushing. However, as the recent international workshop on testing methods concluded, the clinical validation, acceptance and agreement of standardised testing methods by both manufacturers and international bodies would be required.¹⁴ One area of difficulty is with MFP-containing formulations, where analysis has proved challenging and current methods may not have clinical relevance. If these technical difficulties can be overcome, and all manufacturers standardised on this form of expression and did not declare the total, soluble or free ionic fluoride as calculated on a theoretical basis, then it would provide a more meaningful indicator that dental professionals, pharmacies and indeed the consumer could rely upon.

As advocated by Ruth Holt and John Murray, manufacturers should declare the ionic free fluoride concentration of the product. It should be clearly indicated both on the product tube and box. As a simple guide for consumers, this could be done using a symbol system for formulations with less than 1,000 ppm, those in the range of 1,000-1,500 ppm, and those with higher concentrations. However, this should be the concentration available immediately on brushing using internationally agreed methodology, possibly based on the test approved by the ADA,¹⁴ rather than the theoretical amount that might become available post-brushing.

To achieve these aims, there needs to cooperation and agreement between the major toothpaste manufacturers, national consumer and regulatory bodies, and international standards organisations. The international workshop on testing methods has identified some of the problems that need to be addressed and resolved, and in doing so, it has laid a foundation for a forward path. Leadership for this endeavour could be provided by such international bodies as the International Association for Dental Research (IADR), the World Dental Federation (FDI), the European Organisation for Caries Research (ORCA) or the World Health Organisation (WHO).

The current product labelling used by manufacturers to express the effective fluoride concentration in toothpastes is unsatisfactory and is subject to quality control variations in production and storage properties. Cooperation is required between manufacturers, regulatory bodies and standards organisations in order to develop a simple standardised test, which can be used in all production facilities to indicate the free ionic fluoride within the paste after a defined and controlled storage period. If this can be done, combined with a simple labelling scheme on the product, it would provide a more meaningful indicator that dental professionals, pharmacies and indeed the consumer could rely upon.