Sugar-free chewing gum and dental caries: a systematic review.

OBJECTIVE: To appraise existing evidence for a therapeutic / anti-cariogenic effect of sugar-free chewing gum for patients.
METHOD: 9 English and 2 Portuguese databases were searched using English and Portuguese keywords. Relevant articles in English, German, Portuguese and Spanish were included for review. Trials were excluded on lack of randomisation, control group, blinding and baseline data, drop out rate > 33%, no statistical adjustment of baseline differences and no assessment of clinically important outcomes. Reviews were excluded on lack of information, article selection criteria, search strategy followed, search keywords, searched databases or lack of study-by-study critique tables. In cases of multiple reports from the same study, the report covering the longest period was included. Two reviewers independently reviewed and assessed the quality of accepted articles.
RESULTS: Thirty-nine articles were included for review. Thirty were excluded and 9 accepted. Of the 9 accepted, 2 trials of reasonable and good evidence value did not demonstrate any anti-cariogenic effect of sugar-free chewing gum. However, 7 articles, with 1 of strong, and 6 of good evidence value, demonstrated anti-cariogenic effects of chewing Sorbitol, Xylitol or Sorbitol/Xylitol gum. This effect can be ascribed to saliva stimulation through the chewing process, particularly when gum is used immediately after meals; the lack of sucrose and the inability of bacteria to metabolize polyols into acids.
CONCLUSION: The evidence suggests that sugar-free chewing gum has a caries-reducing effect. Further well-designed randomised trials are needed to confirm these findings.

INTRODUCTION

Several publications have suggested that sugar-free chewing gum has an anti-cariogenic effect1,5,8,16,35,41. Such effect is ascribed to the action of 2 factors: saliva stimulation through the chewing process and integration of dietary polyols8.

An increase in stimulated saliva flow has been associated with an increase in plaque pH19 and a higher salivary buffer capacity8. Manning and Edgar25 (1993) reported that chewing sugar-free chewing gum directly after meals reduced the immediate plaque pH response and thus enhanced the potential of enamel remineralisation. Edgar and Geddes9 (1990) suggest further that the anti-cariogenic saliva effect may be further attributed to increased salivary bicarbonate, leading to higher buffer strength, as well as to an increased supply of alkaline substrates to the plaque. Furthermore, an increased salivary flow rate may significantly contribute to oral health through optimized cleansing of the tooth surface and to accelerated clearance of dietary sugars and plaque acids away from the tooth surface9,27.

The most common dietary polyols used in sugar-free chewing gum are Xylitol and Sorbitol8. Most oral bacteria do not metabolise Xylitol and Sorbitol to form acid8. Xylitol is a sugar alcohol derived from pentose sugar xylose and Sorbitol is a sugar alcohol derived from glucose8. Both elicit a gustatory reflex which, together with the chewing process, enhances saliva stimulation8. Unlike Sorbitol, Xylitol has been observed to exhibit a dose-related inhibition of S. mutans' growth in vitro2.

Based on these findings, a caries reduction in patients who chew sugar-free chewing gum is expected. The objective of this systematic literature review was to appraise existing evidence concerning a possible therapeutic / anti-cariogenic effect of sugar-free chewing gum for patients.

MATERIAL AND METHODS

Search strategy

The literature search included 9 English databases: BIOMED CENTRAL, COCHRANE ORAL HEALTH REVIEWS, COCHRANE LIBRARY, DIRECTORY OF OPEN ACCESS JOURNALS, EXPANDED ACADEMIC ASAP PLUS, META REGISTER OF CONTROLLED TRIALS - mRCT, PUBMED, SCIENCE-DIRECT, RESEARCH FINDINGS ELECTRONIC REGISTER – ReFeR and 2 Portuguese databases: BIBLIOGRAFIA BRASILEIRA EM ODONTOLOGIA – BBO, LITERATURA LATINO-AMERICANA E CARIBENHA EM CIÊNCIAS DA SAÚDE – LILACS.

The English keywords "(chewing gum) AND (caries OR tooth decay) and (chewing gum) AND (caries risk)" and Portuguese keywords "(Goma mascar AND cári$) and (Goma mascar) AND (risco cári$)" were used to search the English and Portuguese databases, respectively. The keywords were standardised in both languages for equivalent weight during the literature search. The time-search for publications in all databases was retrospectively unlimited until 09 June 2005. Publications were selected from the search results on the basis that their titles and abstracts were in accordance with the inclusion criteria: (i) relevance to the review objective; (ii) publication in English, German, Portuguese or Spanish. Where only a relevant title without a listed abstract was available, a full copy of the publication was assessed for inclusion. To be selected for review, publications had to fulfill all inclusion criteria.

Publication review

The review followed published guidelines39. All included publications were assessed independently by two reviewers. Disagreements between the reviewers were solved through discussion, until final consent. Only in-vivo and in-situ trials were included. In-vitro studies were judged as providing insufficient evidence for therapy, since they carried the potential error of extrapolation of laboratory results to physiological effects in humans. For that reason they were not included. In cases of multiple reports from the same study, only the report covering the longest period was included. Published trials and literature reviews were assessed according to the exclusion criteria listed in Table 1. Publications were accepted as evidence only if they passed all exclusion criteria.

TABLE 1

Exclusion criteria for trials and literature reviews

Trials	Literature reviews
Lack of randomisation Drop-out rate >33%	Focus on population or intervention not clearly stated in title and abstract
Patients and clinicians not 'blinded' where possible and appropriate	Article methodology describes no clear inclusion and exclusion criteria for reviewed publications
No baseline data provided for both the control and the study group	Article methodology describes no clear search strategy, key words and databases used and includes no study-by-study critique table or discussion of study qualities
	Baseline differences not statistically adjusted
	Clinically important outcomes for patients not assessed.

Criteria for evidence value

The evidence value of accepted articles was rated by application of a structured scoring system3,45 (Table 2). Following a similar validation system used by Zero, et al.45 (2001), included articles were rated as having 'strong', 'good' or 'reasonable' evidence value. The value of evidence depends on the amount of information provided by authors in their articles, to support the methodology used to obtain their results45. Accepted reviews were automatically rated as having strong evidence value. Since publications needed to pass all exclusion criteria in order to be accepted, accepted articles with lowest value ratings were still considered as offering reasonable evidence value.

TABLE 2

Criteria for evidence values

Quality aspect	Criteria	Points
Study setting:	In - situ	1
	In - vivo	2
Article provides information on:	How the sample were collected	1
	How examiners/patients were blinded	1
	How operators were trained or calibrated	1
	Examiners' reliability	1
Sample drop-out rate:	33-20%	0
	10-19%	2
	<10%	3
Followed up period:	<1 year	0
	1 year	1
	>1 year	2

Quality scores (Total number of points): 'Strong evidence' = 10-11 'Good evidence' = 6-9 'Reasonable evidence' = 0-5

RESULTS

After the literature search, 39 articles were found to be in accordance with the inclusion criteria and were selected for review. Of these, 14 were literature reviews and 25 were articles reporting on clinical trials.

Excluded articles

After review, 14 literature reviews and 16 trials were excluded. Most of the excluded reviews1,4,7–9,11,14,29,35,36,38,41,44 were narrative in nature, lacking a stated literature search strategy and lacking stated inclusion and exclusion criteria for literature selection. Most reviews1,4,7,9,11,14,29,36,38,44 also lacked study-by-study critique tables. Five of the reviews7,11,14,36,38 had been published as editorials or short communications and one was a review of 1 single trial42. Five of the 14 reviews only discussed sugar-free chewing gum as part of an overall review theme; such as general chewing gum9,35, dental caries management1,4 and preventive dentistry44.

Of the 16 excluded trials, 11 were not randomised13,15,18,19,21–24,33,37,43; 3 trials5,16,17 had a loss to follow up of more then 33% and 2 trials lacked operator blinding as part of their methodology10,30.

Accepted articles

A total of 9 articles reporting on trials were accepted. The data concerning the accepted articles are shown in Table 3. Two articles20,26 reported on results of 2 trials, each conducted with chewing gum of different polyol content, thus raising the number of accepted individual trials to 11. From these, 8 trials followed in-vivo, and 3 in-situ study designs. The trials reported on the effects of using sugar-free chewing gum with Sorbitol (6 trials), Xylitol (2 trials) and Sorbitol/Xylitol combined (3 trials). The in-situ trials investigated the anti-cariogenic effect of enamel remineralisation on artificial carious lesions and in-vivo trials reported on caries reduction, mainly in permanent teeth (Table 3). The reported use of chewing gum in the accepted trials varied in frequency per day (3 – 7 times) and duration (5-20 min). Five trials reported gum use immediately after meals6,20,28,31,40.

TABLE 3

Main characteristics of the 11 accepted trials

Article/1st Author	Trial Nr	Study design	Dental chewing gum	Application Regime	Control	Caries/Dentition	Follow -up period	Result		Value as evidence
								Anti-cariogenic effect	Reported statistical significance
Creanor, et al.6 (1992)	1	ln-situ	Sorbitol (2.7 g stick)	5x per day, 20 min chewing after meals	Baseline before chewing	Artificial lesion	7 weeks	Enamel remineralisation	(p = 0.07)	Reasonable
Glass12 (1983)	2	ln-vivo	Sorbitol	Daily chewing gum distribution	No gum	Deciduous	2 yrs	Caries increment reduction	(p>0.05)	Good
Machiulskiene et al 20 (2001)	3	ln-vivo	Sorbitol	5x per day, 10 min chewing after meals	No gum	Permanent	3 yrs	DMFS increment reduction	(p<0.05)	Good
	4	ln-vivo	Xylitol		No gum	Permanent	3 yrs	DMFS increment reduction	(p<0.05)	Good
Manning, et al.26(1992)	5	ln-situ	Sorbitol (100%)	After meals for 20 min	Baseline before chewing	Artificial lesion	21 days	Enamel remineralisation	(p<0.05)	Good
	6	ln-situ	Sorbitol/Xylitol (25%/75%)		Baseline before chewing	Artificial lesion	21 days	Enamel remineralisation	(p<0.05)	Good
Möller and Poulsen28 (1973)	7	ln-vivo	Sorbitol (1-2 g)	3x per day, after meals	No gum	Permanent	2 yrs	Carious decay progression reduction	(p<0.05)	Good
Peng, et al.31 (2004)	8	ln-vivo	Sorbitol/Xylitol (55.5%/4.3%)	4x per day, after meals	No gum	Permanent	2 yrs	DMFS increment reduction	(p<0.05)	Good
Petersen, et al.32 (1999)	9	ln-vivo	Sorbitol/Xylitol (55.5%/4.3%)	3 and 5 × per day	No gum/oral health education	Permanent & deciduous	3 yrs	Occlusal DMFS increment reduction	(p<0.01)	Good
Scheinin, et al.34 (1975)	10	ln-vivo	Xylitol	3–7 × per day	Sucrose gum	Permanent	1yr	Caries incident reduction	(p<0.05)	Good
Szöke, et al.40 (2001)	11	ln-vivo	Sorbitol (65%)	After meals	No gum	Permanent	2 yrs	DMFS increment reduction	(p<0.05)	Strong

The results of the in-situ trials were conflicting. Two trials of good evidence value26, 1 with Sorbitol and 1 with Sorbitol/Xylitol chewing gum, reported statistically significant (p<0.05) enamel remineralisation after 21 days, while 1 trial with Sorbitol chewing gum of reasonable evidence value6 showed no significant remineralisation (p = 0.07) after 7 weeks. One in-vivo trial of good evidence value did not demonstrate any significant caries reduction with Sorbitol chewing gum12. All other in-vivo trials, 1 of strong evidence40 and 6 of good evidence value20,28,31,32,34, were able to demonstrate significant caries reduction after chewing of Sorbitol, Xylitol and Sorbitol/Xylitol chewing gum, as compared to no gum use. However, of these trials, 1 trial32 reported such reduction for the occlusal tooth surfaces, only.

DISCUSSION

This systematic review was the first to include English as well as Portuguese databases in its literature search and to review articles in the 4 publication languages: English, German, Portuguese and Spanish, relating to the topic of sugar-free chewing gum and caries. Moreover, the literature search was extended to cover multiple English databases. However, despite this broader approach, some limitations may have affected its results: (i) no hand-searching method was used; thus relevant studies may not have been identified; (ii) used search keywords might not have been broad enough in order to capture all articles listed in the databases. No meta-analysis was undertaken, owing to differences in the methodologies of the accepted trials.

One in-situ6 and 1 in-vivo trial12 out of the 11 separate trials accepted in this review did not report any anti-carious effect of sugar free chewing gum. The reason for not achieving any significant remineralisation (p = 0.07) in-situ after chewing Sorbitol gum 5 times per day for 20 min, as compared to no gum use6, remains unclear. After 7 weeks, remineralisation in the test group was 18.2% and in the control group, 12.1%. Remineralisation in the latter may have been aided by the use of an 1100-ppm-F (NaF) dentifrice given to both groups, thus reducing the effect of Sorbitol gum alone. The in-vivo study by Glass12 (1983) confirmed that Sorbitol chewing gum does not promote tooth decay. However, the results did not confirm any anti-cariogenic effect.

Two in-situ, and 7 in-vivo trials demonstrated significant enamel remineralisation and caries reduction, respectively. In contradiction to the in-situ trial of Creanor, et al.6 (1992), both in-situ trials by Manning, et al.26 (1992), with Sorbitol and Sorbitol/Xylitol chewing gum showed significant remineralisation of enamel. The contradicting outcome of these trials may be due to differences in technical aspects of the investigated artificial lesions. Such aspects may include lesion depth due to length of enamel placement in demineralising solution, as well as differences in enamel composition and structure, which vary between tooth sites. For this reason, the evidence found in our systematic literature review with regard to any significant remineralising effect of sugar-free chewing gum use on carious lesions is inconclusive.

Most of the accepted in-vivo trials showed significant caries reduction. Such reduction appeared to be independent of polyol-type, polyol-composition and concentration and chewing regimes. Machiulskiene, et al.20 (2001) observed no difference between polyol gum and a sugar-free control gum without polyol but a caries-reducing effect of the control gum when compared to no gum use. Petersen, et al.32 (1999) found a caries-reducing effect only for occlusal surfaces and Scheinin, et al.34 (1975) suggested that the anti-cariogenic effects of polyol were due to the lack of its suitability for micro biotic metabolism. Szöke, et al.40 (2001) observed an anti-cariogenic effect after 20 min of chewing sugar-free chewing gum, especially immediately after meals: thus long after all soluble ingredients, such as polyols had been dissolved. It can therefore be concluded that the caries-reducing effect was not due to any therapeutic action of polyol but rather, to the chewing process itself and subsequent saliva stimulation.

According to our criteria, the evidence value of 9 accepted trials was good, while 1 trial was of strong, and 1 of reasonable evidence value. Of these 9 trials, 5 trials6,12,26,28,34 did not elaborate on examiner calibration and reliability in their methodologies. Such lack of information reduced the value of evidence found in this review. The quality and quantity of the found evidence suggests a further need for well-designed randomised trials to further confirm the caries-reducing effect of sugar-free chewing gum, as well as its underlying mechanisms. Further trials should provide data on any remineralising effect of polyol chewing gum and clarify whether specific polyols, such as Xylitol, have any anti-cariogenic, therapeutic effect.

CONCLUSION

The evidence suggests that chewing sugar-free chewing gum, used immediately after meals, reduces caries. The observed caries reduction can be ascribed to saliva stimulation throughout the chewing process, the lack of sucrose and the inability of bacteria to metabolise polyols into acids. No evidence for a direct therapeutic effect caused by Sorbitol or Xylitol was found. Further well-designed randomised trials are needed to confirm theses findings.