| Sign In to gain access to subscriptions and/or personal tools. |
© 2008 SAGE Publications Evidence-based Effectiveness of Topical FluoridesPresented at a symposium entitled "Fluoride and Caries Decline", sponsored by the IADR Cariology Research, Behavioral, Epidemiologic & Health Services Research, and Pharmacology/Therapeutics/Toxicology Groups, presented during the 35th Annual Meeting of the American Association for Dental Research and the 83rd Annual Session of the American Dental Education Association, March 9, 2006, Orlando, Florida, USA, and supported by the Colgate-Palmolive Co.
Barts and The London School of Medicine and Dentistry, Queen Mary, University of London, Turner Street, London E1 2AD, UK; v.marinho{at}qmul.ac.uk
Key Words: Topical fluorides • dental caries • systematic review • meta-analysis • randomized controlled trials Systematic reviews of controlled trials are the gold standard for decision-making and appear at the top of the hierarchy of evidence about effectiveness. They can and should inform decisions about the appropriate use of fluorides to prevent dental caries, and suggest areas for further research. The series of seven Cochrane systematic reviews, published in The Cochrane Library, assesses the effectiveness of fluoride toothpastes, gels, varnishes, and mouthrinses through comparisons against non-fluoride controls, against each other, and against different combinations. Apart from confirming the relative effectiveness of topical fluoride therapy for caries prevention, they address several issues of interest for clinicians and research methodologists. The main features and findings of these reviews are presented here. The paper concludes by considering the evidence and rationale to recommend other sources of fluoride, in addition to fluoride dentifrice, as well as other important implications for research and practice.
I shall begin by attempting to address the importance of systematic reviews in informing decisions about the appropriate use of topical fluorides to prevent dental caries and the research agenda on this topic. The concept of caries prevention with topical fluoride treatments and the animal, laboratory, and human research which created the basis for the widespread use of fluoride in the form of toothpastes, mouthrinses, gels, and varnishes have been extensively documented and discussed. The various topical fluoride interventions have been subjected to intensive clinical testing in randomized controlled trials (RCTs), but the vast majority of reviews on the topic have used traditional narrative methods to summarize this evidence, largely ignoring the levels of evidence available and the variable quality of studies reviewed. It has become clear, however, that information from rigorous systematic reviews that have now been done should inform decisions about population-wide use of fluoride to prevent caries and also about further research. For example, the UK National Health Service (NHS) Centre for Reviews and Dissemination’s (CRD) systematic review of the effects of water fluoridation (McDonagh et al., 2000), the first systematic review undertaken on water fluoridation, which was conducted in an open process and to the highest standards, shows that because research carried out over the past half-century has been of a much lower methodological quality than had previously been reported, there will continue to be a need for the conduct of high-quality studies providing more definite current evidence of the effects, both positive and negative, of water fluoridation. The NHS CRD report points out that the evidence of a benefit of a reduction in caries should be considered together with the evidence of increased prevalence of dental fluorosis. The findings of this review have reinforced the importance of systematic reviews of the large body of experimental evidence on the effects of topical fluorides. The questions being asked relate mainly to the actual effectiveness and the potential risks (mainly fluorosis) that may be expected from the various fluoride-based treatments. This is particularly important in an era of decreased caries prevalence and widespread exposure to fluoride from several fluoride sources.
The main impetus for the growth of systematic reviews of randomized controlled trials (RCTs) in general, and oral health care in particular, is related to the work of The Cochrane Collaboration, an international organization that prepares, maintains, and disseminates rigorous systematic reviews of the effects of health care interventions in all specialties (Clarke and Langhorne, 2001). The Collaboration was named after Archie Cochrane, a British epidemiologist and pioneer in the field of evaluating medical interventions. In his influential book published in 1972, Effectiveness and Efficiency, Cochrane drew attention to the collective ignorance about the effects of health care, made a strong case for the evaluation of new and existing forms of care in controlled trials—especially in the usual case, when the effects of care are not dramatic, but are nevertheless important—and explained how evidence from such trials could help in a more rational use of resources (Cochrane, 1972). In 1979, he said, "It is surely a great criticism of our profession that we have not organised a critical summary, by specialty or subspecialty, adapted periodically, of all relevant randomised controlled trials." The Cochrane Collaboration, established in 1993, is the international collaborative effort in response to Cochrane’s call for systematic, up-to-date reviews of all relevant randomized controlled trials of health care (Chalmers, 1993).
Indeed, the strongest scientific evidence about effectiveness will come from such quantitative study designs that are likely to provide the best means of minimizing systematic errors (or bias), and also to have numerous observations from a randomized body of evidence which is large enough to control for random errors (or chance effects). A variety of study designs has been used to evaluate the effects of care, and a hierarchy of validity exists to grade such studies based on the rigor of their design and their ability to minimize bias (Woolf et al., 1990). The hierarchy of evidence about effectiveness, and variations of the hierarchy proposed and adopted by various evidence-based guidelines organizations worldwide, relates to how confident one can be that the observed effects are attributable to the intervention and are not the result of other factors (Table 1
Although hundreds of reports of randomized trials published since the 1940s helped to bring about a situation in which the effectiveness of topical fluorides in preventing caries had been recognized, systematic reviews of the extensive body of experimental evidence on the effects of topical fluorides were lacking until recently. This is despite the fact that the first reviews using meta-analyses of controlled trials had started to appear in dentistry as early as the mid-1980s—coinciding with the growth in their availability in the field of medicine (Stamm et al., 1984; Clark et al., 1985). Nevertheless, the general picture in the early 1990s was that the effectiveness of topical fluorides had been extensively summarized in traditional narrative reviews based on selected published literature, effectiveness estimates had been reported in broad ranges, and there appeared to be no general agreement on the causes of differences in effectiveness. However, many of the available reviews had highlighted important issues relevant to the assessment of the effectiveness of the various topical fluoride treatments, which started to be formally taken into account in the systematic reviews published subsequently (Johnson, 1993; Helfenstein and Steiner, 1994; Stamm, 1995; van Rijkom et al., 1998; Bartizek et al., 2001; Strohmenger and Brambilla, 2001). In general, the available systematic reviews focused mainly on the evaluation of specific topical fluoride active agents within specific delivery systems, a common characteristic of those published more recently as well (Ammari et al., 2003; Twetman et al., 2003; Petersson et al., 2004; Steiner et al., 2004). The seven Cochrane systematic reviews, which are the focus of this paper, have gone further. Being the most comprehensive to date, they bring together and summarize the large body of knowledge from the available experimental evidence on the main modalities of topical fluoride therapy (TFT) currently used for the prevention of dental caries in children, and examine systematically the main factors that can influence effectiveness (Marinho et al., 2002a,b, 2003a,Marinho et al., b,c, 2004a,Marinho et al., b). The reviews were prepared according to the guidelines of the Cochrane Handbook of Reviews of Interventions (Higgins and Green, 2005), and are published in The Cochrane Library following the editorial process of the Cochrane Oral Health Group (COHG), which is common to all Cochrane reviews.
The Cochrane reviews assess the effectiveness of fluoride toothpastes, gels, varnishes, and mouthrinses through comparisons against non-fluoride controls, against each other, and against different combinations. The first four reviews individually compared fluoride gels, varnishes, rinses, or toothpastes with placebo or no treatment and examined factors potentially influencing effectiveness; the fifth review is a summary of the previous reviews, where additional investigations of covariates across all four sets of controlled trials were performed, and differences between and among fluoride modalities were examined; the sixth review involved head-to-head comparisons of fluoride gels, varnishes, rinses, and toothpastes; and the seventh review involved head-to-head comparisons of these modalities used in combination vs. one form used alone (primarily any topical fluoride plus fluoride toothpaste vs. fluoride toothpaste alone). The reviews were based upon the same comprehensive searches of published and unpublished evidence, with no language restrictions, and collated and critiqued the included trials using similar methodology and measures of effect. One of the advantages of the publication of a series of reviews in this manner, beyond the ability to bring all the evidence together in a consistent way, is that sensible comparisons between and among the different topical fluoride treatments can be made, and results are not interpreted in isolation. The primary outcome used across all reviews was caries increment, and a typical outcome measure, increment in D(M) FS. Topical fluoride treatment effects were expressed mainly by the Prevented Fraction (PF). The PF, more commonly referred to as ‘percentage caries reduction’, is the difference in mean caries increments between the ‘treatment’ and ‘control’ groups, expressed as a percentage of the mean increment in the control group. Decisions about whether trials were included, quality assessment, and data extraction were duplicated in a random sample of one-third of the studies. Major issues considered in the Cochrane reviews were: the potential benefits to be expected from topical fluorides (in terms of the size of the reduction in caries increments that may be obtained from t he single use of the various interventions), how the benefits of topical fluoride therapy may vary according to the influence of potentially important effect modifiers (in terms of the possible dependence of the caries-preventive effects of topical fluorides on background exposure to fluoride sources other than the study options, on baseline caries levels, and on intervention features, such as fluoride content and frequency of use), and whether the benefits differ among the various interventions alone and when used in combination. Potential harmful effects—such as dental fluorosis, oral allergies, tooth staining, or symptoms of acute toxicity—were also considered.
The selection of a topical fluoride intervention should be based on three general considerations: First, the intervention should be effective in preventing dental caries. Second, it should be safe. Last, it should be easy to use and acceptable to the recipient of care. The first question addressed by the Cochrane reviews is, How effective is the use of topical fluorides in the form of toothpaste, mouthrinse, gel, and varnish for the prevention of caries in children when compared with placebo or ‘no treatment’? All four modalities of topical fluorides were found to be effective. The average D(M)FS prevented fractions in the four individual Cochrane reviews ranged from 24% (95% CI, 21% to 28%) for fluoride toothpaste, through 26% (95% CI, 23% to 30%) for mouthrinses and 28% (95% CI, 19% to 37%) for gels, to 46% (95% CI, 30% to 63%) for fluoride varnishes (Table 2  ). In terms of absolute caries reductions per year, in populations with caries increments of around 2 D(M) FS per year, these ranged from 0.46 for fluoride gels to 0.74 for fluoride varnishes (mouthrinses, 0.56; toothpaste, 0.62). However, conclusions on treatment effects were made on a clearer basis in placebo-controlled trials (Table 3), since caries reductions are overestimated in ‘no treatment’ control topical fluoride trials, which are not double-blind and are likely to be of lower methodological quality. This was shown by the results of the summary review (which compiled and analyzed results of the four individual reviews together), and in the fluoride gel review. The pooled D(M)FS prevented fraction for fluoride gel was 21% (95% CI, 14% to 28%), based exclusively on 13 placebo-controlled studies (one trial originally included in the gel review was excluded from the analysis in the summary review because it had a poor estimate of PF)—significantly different from the 38% (95% CI, 0.23 to 0.53) D(M)FS PF for the nine studies that compared the gel with a ‘no treatment’ control group. It is also worth noting that the conclusion that fluoride varnish is effective for caries prevention in primary teeth had not been advanced in previous reviews. The average d(e/m)fs prevented fraction for varnishes was 33% (95% CI, 19% to 48%), based on three trials, two of which used no treatment controls.
While robust evidence on effectiveness was available, the reviews were generally unable to examine the safety of the various topical fluoride interventions, since the trials rarely provided information on fluorosis and other adverse effects. Although this was disappointing, because safety issues should be important determinants in both public health and individual clinical decisions, the lack of direct evidence on the risk of adverse effects in the topical fluoride reviews was not unexpected, since problems with using randomized trials to assess and report harmful effects have been found in health care generally (Cuervo and Clarke, 2003). Other sources of evidence on harm, such as observational studies, might have been considered, but these would have problems of their own, since it is difficult to know how many people have been exposed, and the susceptibility to bias is considerably greater. Relative to acceptance, an interesting finding from the review comparing topical fluorides against each other was that if children were allocated to fluoride toothpaste groups, they were more likely to stay in the study than if they were given alternative forms of topical fluorides (evidence of acceptability assessed through differential dropout rates).
Another major question addressed in the Cochrane reviews is whether the caries-preventive effect is influenced by initial level of caries, background exposure to other fluoride sources, use under supervision, fluoride concentration and frequency of use, and the form (modality) of topical fluoride adopted. Most of the factors potentially influencing effectiveness were formally examined in the individual reviews. The largest number of studies reporting relevant data was in the toothpaste review, which made it possible to carry out such investigations more reliably in this particular review, which included 70 placebo-controlled trials in the meta-analysis, and across all topical fluorides in the summary review, which included 133 trials. In the fluoride toothpaste review, higher D(M)FS prevented fractions were significantly associated with higher initial levels of caries, higher fluoride concentration, higher frequency of use, and supervised brushing; however, no association between treatment effect and exposure to fluorides from other sources was found in this and in the other reviews—estimates of topical fluoride treatment effect were similar between trials conducted in fluoridated and non-fluoridated areas (Table 4
Analysis across reviews (in the summary review, fifth in the series) also showed a significant influence of initial level of caries, supervision of self-applied topical fluoride use, and topical fluoride modality on the prevented fraction. For the influence of fluoride modalities, results suggested no significant differences in treatment effects among fluoride gel, mouthrinse, and toothpaste, but significantly lower D(M)FS prevented fractions for fluoride gel, mouthrinse, or toothpaste in comparison with fluoride varnish. However, in these adjusted indirect comparisons of all four fluoride modalities, it is difficult to rule out the possibility of an overestimation of the size of the differential effect in favor of fluoride varnish (14% on average), since relatively few varnish trials were included, and few among these were placebo-controlled trials. The question on the effectiveness of one type of topical fluoride compared with another was also addressed by direct head-to-head comparisons between and among fluoride modalities, in the sixth review in the series. There was, however, a relatively small number of trials for each direct comparison and a general lack of statistical significance of the results for virtually all analyses. Nevertheless, results from the nine trials comparing fluoride toothpaste with either gel or mouthrinse were consistent with no evidence of an important differential effect. The question addressed by the final review relates to the effectiveness of the simultaneous use of more than one topical fluoride therapy compared with the use of one topical fluoride alone. There was indication of a greater caries-inhibiting effect with the combined use of topical fluorides in the permanent dentition for most of the relevant comparisons. But with a general lack of randomized trial evidence for most comparisons, a modest treatment effect may have been missed. There was, however, evidence from nine trials showing that the simultaneous use of a topical fluoride treatment with fluoride toothpaste results in an enhanced caries-inhibiting effect compared with the use of toothpaste alone—on average, an additional 10% (95% CI, 2% to 17%) reduction in D(M)FS can be expected. To what extent the statistically significant caries reductions on the order of 10% should be considered important is a decision that requires consideration of other relevant aspects. It should be noted, however, that when two methods of applying topical fluoride are already in use in a population, additional benefits may be small, especially when the measures act identically or by similar mechanisms.
The evidence on the beneficial effects of topical fluorides is consistent and strong, based on a sizable body of evidence from randomized controlled trials. The research involving more than 65,000 children and adolescents in over 130 controlled trials compiled in the Cochrane reviews shows that fluoride toothpastes, mouthrinses, gels, and varnishes can reduce dental caries, regardless of water fluoridation or other sources of fluoride exposure. It also shows that the caries-preventive effect of fluoride toothpaste increases when there are higher initial levels of D(M)FS and when a higher fluoride concentration/frequency of application is used, and that supervising a child’s use of the fluoride (toothpaste or mouthrinse) leads to greater benefits. The size of the reductions in caries increment in both the permanent and the deciduous dentition emphasizes the importance of including topical fluorides delivered through toothpastes, rinses, gels, or varnishes in any caries-preventive program. It was noted that fluoride toothpaste—the most readily available form of fluoride, which is commonly linked to the decline in caries prevalence in many developed countries—can protect children and adolescents against dental caries as much as other topical fluorides, and that young people are more likely to persist with using toothpaste than with using fluoride mouthrinses, or having gels or varnishes applied. The higher acceptability of toothpaste makes its regular use more likely, thereby improving effectiveness. This is an additional indication for the major role of fluoride toothpaste as an effective and acceptable public health approach for the prevention of dental caries. Such evidence was considered by the WHO in its recommendation that every effort must be made to develop affordable fluoridated toothpastes for use in developing countries (Jones et al., 2005). The evidence from the systematic reviews also shows that children using another form of topical fluoride therapy (TFT) with fluoride toothpaste will experience additional reductions in dental caries, compared with children using fluoride toothpaste only. The size of the caries-preventive effect may not be substantial, and because other important outcomes, such as possible side-effects from the combined use of topical fluorides and toothpaste, were not addressed in the trials, it is more difficult to derive clear recommendations on the benefits of using another topical fluoride in addition to toothpaste. However, current clinical practice usually includes an additional topical fluoride modality (over toothpaste) for children at higher risk of developing dental caries. Since increased effectiveness of topical fluorides is to be expected in children with higher initial D(M)FS scores, this practice may be considered in populations with a caries increment of around 2 D(M)FS per year or more. It should be recognized, however, that such an approach reinforces targeting preventive care to high-risk sub-populations. That high-risk approach fails to deal with the majority of new caries, which occur in the population who are at lower risk (Batchelor and Sheiham, 2006). Thus, the aforementioned caveat should be considered before application of the evidence related to the question on whether to use an additional topical fluoride (with toothpaste). The general lack of information across the topical fluoride reviews on relevant outcomes other than caries increment also makes it more important that further experimental research on topical fluorides include assessments of potential benefits as well as harms, especially since there is epidemiological evidence linking fluorosis with exposure to topical fluorides (Maupome et al., 2003; Bottenberg et al., 2004; Harding et al., 2005). The Cochrane topical fluoride reviews are now being updated, as evidence from new trials is being incorporated into the existing reviews. It is expected that the reporting of adverse effects in new trials might be generally improved through initiatives such as the CONSORT for randomized trials, and the QUOROM for reviews (Moher et al., 1999, 2001), as these become more widely adopted by dental journals. This would improve the future identification and quantification of potential harmful effects, and acceptability, of topical fluorides.
No funding from commercial entities has been received at the time of the production and publication in The Cochrane Library of the Cochrane reviews cited in this paper, and these reviews remain independent throughout and free from any potential conflict of interest.
Advances in Dental Research, Vol. 20, No. 1,
3-7 (2008)
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
Introduction
Archie Cochrane, The Cochrane...