1 The effect of non-cariogenic sweeteners on the prevention of dental caries: A review of the evidence
Catherine Hayes, D.M.D., D.M.Sc.
Harvard School of Dental Medicine
Department of Oral Health Policy and Epidemiology
188 Longwood Avenue
Boston, MA 02115
The complete version of this paper can be viewed at:
The role of sugar substitutes such as xylitol and sorbitol in the prevention of dental caries has
been investigated in several clinical studies. The purpose of this report is to review the current
published evidence regarding the relationship between sugar substitutes and dental caries. A
literature search was conducted using MEDLINE and EMBASE and included studies published
from 1966-2001. Studies, which included human subjects and were published in English were
included in this review. A total of 14 clinical studies were reviewed which evaluated the effect of
sorbitol or xylitol or the combination of both sugar substitutes on the incidence of dental caries.
Most of the reports were of studies conducted with children outside of the United States. These
studies demonstrated a consistent decrease in dental caries, ranging from 30-60 percent, among
subjects using sugar substitutes as compared to subjects in a control group. These caries rate
reductions were observed in subjects using xylitol or sorbitol as the sugar substitute in chewing
gum or toothpaste. The highest caries reductions were observed in subjects using xylitol. These
findings suggest that the replacement of sucrose with sorbitol and xylitol may significantly
decrease the incidence of dental caries. Keywords:
dental caries, systematic review, xylitol, sorbitol, sugar substitutes, prevention
The role of sucrose and other fermentable carbohydrates in the etiology of dental caries
has been well established. Since it is known that sugared chewing gums may increase the risk of
dental caries, it has been proposed that the replacement of sucrose in chewing gum or candies
with a sugar substitute such as xylitol, may contribute to caries prevention.
This relationship has
been studied in-situ and in several clinical studies,1-3 that will be the focus of this report.
It is believed that the benefits of sugar-free gums may be two-fold; 4-6 1.) decreased lactic
acid production and increased salivary flow potentially leading to an increased buffering of acids
in plaque and 2) increased supersaturation of saliva with the mineral ions as well as enhanced
clearance of sugars from the mouth. Thus sugar substitution and salivary stimulation could, it has
been argued, be equally responsible for the non-cariogenicity of sugar-free chewing gum.6
Sorbitol and xylitol are the most commonly used sugar substitutes. Although sorbitol is
metabolized at a slower rate than sucrose and not at all by most microorganisms, it can be
fermented at a slow rate by all of the mutans streptococci including S. mutans
while xylitol is
considered to be non-acidogenic.7-10 REVIEW OF PUBLISHED STUDIES
In a clinical trial, in Puerto Rico, 2,601 schoolchildren were randomly assigned to two
study groups.11 One examiner evaluated caries in the permanent dentition at baseline and after
two and three years of follow-up with the use of artificial light, mirror, explorer and radiographs.
After baseline exams classrooms were randomly assigned to receive no gum or sugar-free
chewing gum (Extra Orbit, Wrigley) with sorbitol (40-60%), mannitol (4-15%) and aspartame
(<0.6%) sweeteners. Children were instructed to chew three times per day for 20 minutes. The
change in DMFS index was evaluated adjusting for age, sex, baseline scores (DMFS) and
baseline surfaces at risk, school, treatment by classroom, and within treatment by school. The
mean age was 11.65 years in the control group and 11.72 years in the treatment group. Subjects
in the chewing gum group had 6.4 percent fewer new DMF surfaces than controls. These
differences were statistically significant. It is possible, however, that these results are biased
since it is impossible to know how often the children in the "no gum" group actually chewed
gum with or without sucrose outside of school. It is possible that they did use chewing gum
outside of the school thus increasing their risk of developing caries.
In a study within the VA system, patients were enrolled in a double-blind randomized
clinical trial as part of standard recall visits.12 All inpatients with exposed root surfaces were
invited to participate. One hundred and eighty-eight consented to participate and were
systematically assigned to xylitol or sorbitol and followed for 1.8 years. There were 40 subjects
in each of the intervention groups and 105 in the non-participating group. There were 36 root
surface caries among 2,632 person years of risk in the sorbitol group and six lesions among
2,349 person years of risk in the xylitol group resulting in a relative risk comparing the two types
of gum of 0.19 (0.06,0.62) (p=0.0065) indicating a significant reduction in carious lesions on
exposed root surfaces among those who chewed xylitol gum. An adjustment by age, gender,
number of carious root surfaces at baseline, or caries-free root surfaces at baseline, did not
appreciably alter these results. The data were not analyzed against the non-participating group
since the authors believed such an analysis to be "unjustified.”
In a double-blind cohort study in Belize, 1,277 schoolchildren (mean age 10.2yrs) were
randomly assigned (by school) into nine treatment groups: four xylitol groups (4.3-9.0g/day);
two xylitol-sorbitol groups ( 8.0-9.7g/day); one sucrose group 9.0g/day; one sorbitol group
(9.0g/day); and one sucrose group.13 All participants were fourth grade pupils attending 19
public schools. The gums used in the study were packed in number coded blank wraps to ensure
that students were blinded as to the type of gum. During each of the approximately 200
schooldays per year there were five minute periods of gum use supervised by teachers with
timers. Verbal and written instructions were provided during school vacations.
The onset of a carious lesion on a previously sound or unerupted tooth was evaluated as
the outcome. A total of four blinded and calibrated dentist examiners carried out the exams at
baseline, 16, 28, 40 months, applying the WHO criteria for caries detection. A carious lesion was
recorded if physical discontinuity and softness of the enamel in either pits or smooth surfaces
were evident. Radiographs were not used to assess caries incidence and enamel-only caries were
scored as D0. The overall loss to follow-up was 32 percent with an uneven distribution between
groups. The largest reduction in caries rates occurred in xylitol groups that was significant when
compared with sorbitol or sucrose. Relative risks for caries rates were adjusted for age, gender,
DMFS and number of sound surfaces at baseline. The most significant caries reduction was
observed in the group assigned to the highest xylitol concentration (RR = 0.27) (0.20, 0.36). The
protective effect of xylitol increased with increasing xylitol composition. The group assigned to
the sucrose chewing gum exhibited a slight increase in caries rate (RR = 1.20) (0.96-1.49) that
was not statistically significant. Sorbitol decreased caries rates significantly (RR= 0.74) (0.60-
0.92) as did the sorbitol/xylitol groups.
Another study in Belize evaluated the effect of xylitol and sorbitol chewing gums on
caries rates in primary teeth with six year old subjects. This study demonstrated a lower rate of
caries in subjects in the xylitol or sorbitol pellet groups compared to a group of children who
were not assigned to a chewing group, with relative risks reported as 0.35 (.21-.59) and .44 (.30-
Subjects from the two cohort studies in Belize were evaluated to determine the effect of
xylitol on rehardening or non-progression of carious lesions.15 The rehardening of dentinal caries
was examined by blinded examiners with explorers and fiber optic lights using the same criteria
as in the original studies. Radiographs were not used. The number of lesions that went from D3
to D0 or D4 to D0 were recorded for each of the nine groups. The following formula was used to
assess the rate of caries arrest and nonprogression: # surfaces which were diagnosed to reharden or non-progressed carious surfaces with a caries diagnosis of D3 or D4 at baseline
Arrest or non-progression of caries was seen more frequently in subjects using the xylitol
gum. The group with the highest percentage of xylitol exhibited a higher percent of arrested
carious lesions (27%), compared to the no gum group (9%) or the sorbitol group(7%) (p < 0.05).
A five year follow-up study of the effect of xylitol candies or gums was conducted with 740 10
year-old children in 12 schools in four towns in Estonia.16 The candies were used for two years
and the gum for three years. Two examiners conducted blinded exams at a local school dentist’s
office with mirror and explorer using the WHO criteria. After three years 75 percent of the
original group was re-examined. Both xylitol groups had significantly reduced caries rates
compared to controls. The mean DMFS scores after three years, adjusted for age, gender,
examiner and baseline DMFS were: 4.42 (+4.36) in the control group, 1.87 (+2.55) in the
chewing gum group and 2.77 (+3.05) and 1.72 (+ 2.04) in the two candy groups. The overall
reduction in caries rates compared to controls was 53.5 percent in the chewing gum group and
33-59 percent in the two candy groups compared to controls. These results were statistically
significant (p < 0.005).
A demonstration project in Madagascar in which school children were randomly assigned
to polyol chewing gum or control group included children in grades one and four in six
schools.17 All children received a school-based oral health education program that included daily
supervised toothbruhsing. The test group also received a chewing gum that contained 55.5
percent sorbitol, 4.3 percent xylitol and 2.3 percent carbamide that they received 3-5 times per
day. Dental examinations were performed by three calibrated dentists at baseline and after three
years of follow-up using a standard explorer, mouth mirror and daylight. After three years of
follow-up the overall DMFS scores did not differ significantly among any of the study groups.
The only statistically significant finding was a decrease in occlusal caries in children in grade
one in the xylitol group. It is interesting to note that the findings were different than the other
studies cited in this review. The daily supervised toothbrushing that all children participated in
may have made the groups more similar in terms of oral hygiene status thus reducing the
detectable difference in caries rates.
The discussed studies evaluated the use of chewing gum or candies with sugar
substitutes on caries rates. One study evaluated the caries inhibitory effect of xylitol in a
dentifrice.18 This study was conducted in Costa Rica beginning in 1987 with 2,630 school
children aged eight to ten. A calibrated dentist conducted clinical evaluations on all children at
three time points throughout the study. The children were divided into two groups, 10.243
percent NaF/silica dentifrice or a dentifrice containing 0.243 percent NaF/silica plus 10 percent
xylitol, and stratified by age and sex. They brushed twice daily with the study toothpaste
including once daily at school and home during weekdays and twice daily on the weekends.
The DFS incremental change was 3.3 in the control dentifrice and 3.1 in the dentifrice
containing xylitol, representing a 9.1percent difference in caries incremental change (p<0.01).
After three years the loss to follow-up was 36 percent that was consistent between the two
groups. The mean DFS changes from baseline were 5.7 for the control group and 5.0 for the test
group (p < 0.001). LONG-TERM EFFECTS
The long-term effects of sugar-free gum chewing have been reported in a single study in
which children were re-examined five years after a two-year gum chewing study ended.
Comparisons were made between sorbitol, xylitol and no gum chewing. The sorbitol gum did not
have a significant long-term effect on caries reduction. The xylitol and xylitol/sorbitol groups
demonstrated significant long-term caries reductions with relative risks of 0.41 (0.23,0.75) and
0.56 (.36,.89) respectively. The protective effect of xylitol depended on when teeth erupted.
Teeth erupting after one year of gum chewing or after the two year period had ended,
demonstrated the most significant long-term caries reductions (93% and 88% respectively).19 Streptococcus Mutans
The effect of sugar substitutes on changes in S. mutans
levels also have been
investigated. All studies have consistently demonstrated that xylitol use did significantly reduce
the levels of S. mutans
The effect of sugar substitutes on changes in caries rates has been evaluated in several
observational studies as well as clinical trials with results consistently demonstrating a protective
effect of xylitol on caries incidence. Sorbitol also was shown to decrease caries rates compared
to controls, however, the reductions in caries rates were greatest when xylitol was the sugar
substitute. Some limitations of previous studies include the lack of radiographs in caries
diagnosis, high loss to follow-up, potential confounding and bias due to nature of long-term
community intervention studies.
The criteria for causality: consistency, strength association, biologic plausibility,
temporal sequence and dose response relationship should be considered. First, these studies are
remarkably consistent both in terms of the magnitude of the effect observed as well as the
consistent demonstration of the superiority of xylitol compared to sorbitol in decreasing the risk
of dental caries. Second, the relative risks observed, 0.19-0.4, are considered strong evidence of
a protective effect. Third, it is biologically plausible that xylitol can reduce dental caries since
the pH of plaque is not lowered to the range that would increase caries risk with xylitol
compared to sucrose. Fourth, a dose response trend was observed in the two studies that
evaluated different concentrations of xylitol with the greatest effect observed in the subjects
using the strongest xylitol preparations. Although several of these studies were flawed it is
unlikely that future studies can improve on what has been done to date. Furthermore, since the
evidence suggests a strong caries protective effect of xylitol it would be unethical to deprive
subjects of its potential benefits. Given that several of the criteria for causality are met, it is
concluded that xylitol can significantly decrease the incidence of dental caries.
1. Creanor SL, Strang R, Gilmour W, et al. The effect of chewing gum use on in-situ enamel
lesion remineralization. J Dent Res 1992;71:1895-1900.
2. Leach SA, Lee GTR, Degar WM. Remineralization of artificial caries-like lesions in human
enamel in-situ by chewing sorbitol gum. J Dent Res 1989;68:1064-8.
3. Manning RH, Edgar WM, Agalamanyi EA. Effects of chewing gums sweetened with sorbitol
or a sorbitol/xylitol mixture on the remineralization of human enamel lesions in-situ. Caries Res
4. Edgar WM, Higham SM, Manning RH. Saliva stimulation and caries prevention. Adv Dent
5. Edgar WM. Sugar substitutes, chewing gum and dental caries - A review. Br Dent J
6. Edgar WM, Geddes DAM. Chewing gum and dental health - A review. Br. Dent J
7. Gales MA, Nguyen TM. Sorbitol compared with xylitol in the prevention of dental caries.
Ann Pharmacother 2000;34:98-100.
8. Hogg SD, Rugg-Gunn AJ. Can the oral flora adapt to sorbitol? J Dent 1991;19:263-71.
9. Makinen KK, Makinen PL, Pape HR, et al. Conclusion and review of the “Michigan Xylitol
Programme” (1986-1995) for the prevention of dental caries. Int Dent J 1996;46:22-34.
10. Rateitshak-Pluss EM, Guggenheim B. Effects of a carbohydrate free diet and sugar
substitutes on dental plaque accumulation. J Clin Periodontol 1982;9:239-51.