The Effect of Chewing Gum Use on in situ Enamel Lesion Remineralization S.L. CREANOR, R. STRANG3, W.H. GILMOUR1, R.H. FOYE, J. BROWN2, D.A.M. GEDDES, and A.F. HALL Oral Biology Group, 'Department of Public Health and 2Department of Prosthodontics, University of Glasgow Dental School, 378 Sauchiehall Street, Glasgow G2 3JZ, and 3Department of Clinical Physics & BioEngineering, Greater Glasgow Health Board, 11 West Graham Street, Glasgow G4 9LF, Scotland, United Kingdom Two independent cross-over studies investigated the possibility of enhanced early enamel lesion remineralization with the use of chewing gum. The first study involved a sorbitol-containing chewing gum, and the second, which had an identical protocol, tested a sucrose-containing chewing gum. In each study, 12 volunteers wore in situ appliances on which were mounted enamel sections containing artificial caries lesions. Subjects brushed twice daily for two min with a 1100-ppm-F (NaF) dentifrice (control and test) and in the test phase chewed five sticks of gum per day for 20 min after meals and snacks. Microradiographs of the enamel lesions were made at baseline and at the end of the seven-week experimental period. In the sugar-free gum study, the weighted mean total mineral loss (Az) difference [(wk7 - wkO) x (-1)] was 788 vol.% min. x pum for the gum, corresponding to remineralization of 18.2%, vs. the control value of 526 vol.% min. x jum, 12.1% remineralization (p = 0.07). There were no significant differences for the surface-zone (p = 0.20) and lesion-body (p = 0.28) values. In the sucrose-containing gum study, the Az difference was 743 vol.% min. x pn for the gum, corresponding to a remineralization of 18.3%, vs. the control value of 438 vol.% min. x pm, 10.8% remineralization (p = 0.08). The surface-zone values were not significantly different (p = 0.55). For the lesion body, however, the sucrose-containing gum value of 6.11 vol.% min. was significantly different (p = 0.01) from that of the control (2.81 vol.% min.). J Dent Res 71(12):1895-1900, December, 1992
Introduction. The effects of chewing gum use on oral health were reviewed recently by Edgar and Geddes (1990). In evaluations of possible effects of chewing gum on caries, the usage pattern of the gum has to be considered. Edgar et al. (1975), using a five-minute chewing period, investigated the effects of gums alone and found that sorbitol-based chewing gums had little, if any, effect on plaque pH, while sucrose-containing gums lowered plaque pH. Park et al. (1990) investigated the effect of chewing a sorbitol gum for 10 min after the ingestion of snack foods. They found that use of chewing gum resulted in a smaller fall in pH and a quicker rise to resting pH levels. This they attributed to an increase in salivary flow and buffering capacity of the saliva. Other studies have looked at longer chewing periods. Jensen and Wefel (1989) and Manning and Edgar (1991) found that the chewing of sugar-free gums for 20 min after an acidogenic challenge also decreased the time for the resting pH to be regained. Importantly, in both these studies, sucrose-containing gums were found to have the same effect, which were not statistically different from those of the sugar-free gums. There have been few in situ de/remineralization chewing gum studies published. Leach et al. (1989) reported that chewing a sorbitol gum for 20 min after meals and snacks resulted in Received for publication March 13, 1992 Accepted for publication May 27, 1992 This study was supported by a grant from the Wm. Wrigley, Jr. Company (Chicago, USA).
significantly greater remineralization. The aim of this in situ study was to test the hypothesis that chewing either sorbitol- or sucrose-containing gum for 20 min after eating would result in increased remineralization of artificial enamel lesions beyond that which could be obtained with a fluoridated dentifrice alone.
Materials and methods. Two cross-over tests were run separately and were designed to investigate the possibility of enhanced early enamel lesion remineralization with use of chewing gum. The first study was carried out with a sorbitol-containing gum, and the second, which had an identical protocol, tested a sucrose-containing chewing gum. Subjects and experimental protocols.-Twelve healthy adult subjects participated in each study, which had been approved by the local Ethics Committee. Ten subjects were common to both studies. Upper and lower alginate impressions were taken of the subjects' dentitions, from which plaster casts were constructed. Lower appliances, similar to those described by Creanor et al. (1986), were constructed for each subject. These appliances are worn on the lingual aspect of the mandibular arch. Enamel sections are mounted in recessed areas on the fitting surface of the device, positioned just below the lingual aspect of the lower first and second molar teeth (Fig. 1). The volunteers were provided with the devices (without specimens at this point) prior to being tested, to be worn for a few days so that any irritating areas could be adjusted. All subjects were provided with an ample supply of 1100-ppm-F (NaF) dentifrice at least four weeks before the studies. This dentifrice was to be used throughout the seven-week period of both studies, and even when the appliance was not being worn (i.e., between the tests). Volunteers were instructed to brush twice daily (morning and night) for two min, with a pea-sized amount of toothpaste on a Sensodyne 3.5 toothbrush. Those wishing to brush more frequently were supplied with a non-fluoride dentifrice (1.1 ppm). Volunteers were requested not to rinse at the end of the procedure. During the period of the tests, no other fluoride supplement was to be used. Appliances were worn at all times, including during sleeping and eating, apart from a short period to allow for the cleansing of the lingual aspects of the lower natural dentitions. In an attempt to standardize the diet in both studies, we recorded a weekly record ofeach subject's food intake for the first week of each test. A copy was made and the sheet returned to each subject, with the request that he/she should keep to a similar diet throughout both test and control phases of each study. Volunteers were provided with a supply of snack foods, one of which was to be consumed each day at mid-morning and at mid-afternoon as a snack. No other dietary restrictions were imposed. During the first phase of each ofthe studies, six subjects were instructed to chew five sticks of gum per day, each for 20 min, immediately after breakfast, lunch, and dinner, and after each of the snacks. The remaining six subjects acted as controls, but followed the gum protocol during the cross-over phase, while the first six carried out the control portion. For both tests, the gum was supplied by the Wm. Wrigley, Jr. Co. (USA), and for the first 1895
CREANOR
1896
et
J Dent Res December 1992
al.,
Az difference
vol
min
pm)
x
-
2200
2000'
+4
oniroli
Team
1800
16001400-
12001000
P -t
800
600400200 Z%
!
0
"t 4
-200
0
Fig. 1--The in situ appliance, showing the recessed trough areas on the fitting surfface of the device, positioned just opposite the lower first and second molars. Two enamel specimens are mounted on the base of each
troughl
US), This product weighs 2.7 gper stick, sweetener, along with trace amounts of hydrogenated glucose syrup and aspartame The gum for the test
was
Orbit (Extra
with sorbitol
second test
in
the main
as
was
Doublemint,
which contains 2.5 g of
(sucrose, glucose, glucose syrup)
per
Sectionm preparation.-Artificial enamel lesions were created in sound human premolars extracted for orthodontic reasons. Teeth were coated with an acid-resistant nail polish
on
all aspects, apart from four to six windows (each 1 x 10 their buccal Each tooth was immersed for five
in
approximately
mmol/L
ing 2.0
a
demineralization solution
phosphate, and at a pH of ppmn fluoride and was changed
daily. a
12 longitudinal sections were cut 'rom
Microslice 11 Hard Tissue Diamond Saw to
a
each
tooth
thickness
approximately 250 pm; these were hand-ground, with a alumina slurry, to a final measured thickness of between
of
and 140 pm.
(As).
Baseline mineral content measurements
source-specimen distance of 300
mm.
Mineral
al., 1987). Measurements were made of the total mineral loss (As) and the volume percent mineral contents of the surface zone and lesion body. The differences (week 7 week in As values
0)
sitometric
microden multiplied by that, parameters, negative values represent demmneraliza-
tion and
one so
total min-
that thc
amount of
for all
positive values remnineralization within the lesion. -For each lesion, the difference between
Statistical analysis
the baseline value (week 0) and the final value (week 7)
was
such variations, plots of the differences against riade for the lesions from each volunteer
control phases.
These
plots
Regression lines through ifhese plots of thie predicted change
nation by
were
made for As,
and LB.
use
of
baseline
common
a
value,
the
determination
Pt
baseline
nean
of all
of predicted
the test and control
for volunteer
control (P
profiles through the lesions were calculated from a microdensitometric analysis of the microradiographs tAngmar et al., 1963). Microdensitometry was performed with a Leitz ASBA image analyzer connected to a microcomputer (Strang et
minus
microdensitometric parameters: SZ), and lesiorn-body (LB)
were
for the test and
120
were
content
have all been
three
e
shown (Strang et at. 1987)
values
for the test
Microradiograplhy
a
th
(As), surfaicesone
To allow for
baseline
fine
made, and the sections were then varnished on all cut aspects apart from their natural enamel surfaces, At the end of the seven-week period, the sections were detached from the appliances, the varnish removed, and microradiographs repeated. Four sections were allocated to every appliance on each of the test and control phases for both studies. niicrodensitonmetry. -Mi croradiographs of the enamel sections and an aluminum stepwedge were taken, at baseline and at the end of the seven-week experimental period, with a 20-minute exposure on Kodak High Resolution plates (Type IA) by a Cu(Ka-) x-ray source operating at 20 kV and 30 mA at
for
contents. It has been
for
From ten to with
9000
eral
days
contamn-
calcium, 2.0 mmnol/L
4.5. The solution contained 0.01
calculated
mineral loss
mm) SZ,
surfaces.
50 mL of
8000
7000
remineralisation is related to baseline lesion demineralization
in vitro
on
.3000 4000 6000 6000
2000
Baseline Az (% vol minx pm) Fig. 2 Plot of Az diffrence [wkl w kO) x -11 against baseline Az fPr each lesinacr volunteer (A) fnr both the suicrose-containing gum and control rurns. Pt predicted test (gum) and P. predicted control values for thtvolunteer
sugars
3.2-g stick.
1000
A, Fig.
phases.
were
used for determi"
(wk7 -wk0) for each volunteer, value (see Fig. 2). This same lesiorns
in
each study,
was
used
charges for all volunteers for both, Thus fbr sucrose-containing gum
2 indicates that his
predicted change
(Pt) is approximately 810 vol.% min
)is approximately 230 vol.% min.
x
x
pmn and
ptm
in As
for the
These values
reported in the first row of Table 2. A weighted analysis of variance was undertaken with use of these predicted changes. Weighted analysis was used since these predicted changes are based on different numbers of lesions and are, therefore, estimated with differing degrees of precision for each subject and treatment. It was decided that the predicted changes should be weighted by use of the square root of the number of lesions for each subject/treatment. This analysis Was performed with tihe MINITAB statistical package with the general linear models command (GLM). In addition, since of
and
there
was
P.
then
are
some evidence
in both studies of
a
greater diffrence
in favor of the test (i e.,
who used the
gum in the first
was
in the
gum) for those volunteers experimental run, a period effct
included
analyses.
Since
prexvious
difference would
work with gums
be
would
suggest
that ary
in favor of the gum, one-tailed tests
were
used.
Results.
Sigarrfre0gu.The mean baseline values for all lesions (n= 264) in the sugar-free gum study (ie. both gum and control lesions) were: Az, 4332 (vol/v min x jim); surface zone (SZ), 62.0 (vol.% min.); and lesior body (LB) 42.1 (voA min.). The predicted changes in the three parameters are shown in Table
Vol. 71 No. 12
In situ REMINERALIZATION WITH CHEWING GUM
1897
TABLE 1 PREDICTED CHANGES, FOR EACH VOLUNTEER, IN (a) Az, (b) SURFACE-ZONE (SZ) MINERAL CONTENT, AND (c) LESIONBODY (LB) MINERAL CONTENT FOR TEST AND CONTROL (sugar-free gum; n = number of lesions) Test
Subject
n
Control
Change
A B C D E F G H I J K L
16 16 16 11 12 12 16 6 8 12 9 3
15.6 14.9 10.9 1.1 2.5 14.4 4.9 0.2 5.0 4.1 9.0 4.3
A B C D E F G H I J K L
16 16 16 11 12 12 16 6 8 12 9 3
5.7 4.7 5.7 0.7 1.3 7.7 0.2 5.7 1.6 0.8 1.6 -3.8
A B C D E
16 16 16 11 12 12 16 6 8 12 9
12.9 11.8 10.8 0.5 2.0 20.9 4.3 0.1 6.4 4.0 8.6
3
10.8
F
G H I J K L
Change (a) Az (vol.% min. x pm, x 100) 14 4.2 10 3.3 13 7.1 18 6.7 17 -4.1 5 5.4 9 5.0 9 12.8 6 7.2 6 0.2 16 10.8 n
12.1 (b) Surface-zone mineral content (vol.% min.) 14 1.3 10 -2.7 13 2.5 18 2.4 17 -2.0 5 1.5 9 4.2 9 5.2 6 2.3 6 -5.0 16 4.9 4 3.5 (c) Lesion Body (vol.% min.) 14 3.2 3.8 10 5.1 13 9.1 18 2.3 17 5 4.2 9 2.1 9.5 9 10.1 6 1.8 6 10.4 16 4
4
9.8
Difference
95% C.I. for Difference
11.5 11.6 3.7 -5.6 6.6 9.0 -0.2 -12.6 -2.2 3.9 -1.8 -7.7
8.7 8.2 -0.8 -9.1 2.7 1.3 -2.6 -28.2 -7.2 0.1 -5.3 -35.2
to to to to to to to to to to to to
14.2 15.1 8.2 -2.0 10.6 16.6 2.3 3.0 2.8 7.8 1.8 19.7
4.4 7.4 3.2 -1.7 3.3 6.2 -4.1 0.5 -0.8 5.8 -3.3 -7.4
2.8 3.3 1.1 -4.3 0.9 1.3 -7.9 -12.1 -2.9 -0.1 -6.2 -57.2
to to to to to to to to to to to to
6.0 11.5 5.3 1.0 5.7 11.1 -0.2 13.2 1.4 11.8 -0.3 42.5
9.8 8.0 5.8 -8.7 -0.3 16.7 2.3 -9.4 -3.7 2.2 -1.8
6.4 3.3 2.4 -11.7 -3.2 13.0 0.5 -14.6 -6.8 -1.0 -4.1
to to to to to to to to to to to
13.1 12.7 9.1 -5.6 2.6
1.0
20.4
4.0 -4.2 -0.7 5.3 0.6
-8.0 to 10.1
1898
J Dent Res December 1992
CREANOR et al.
TABLE 2 PREDICTED CHANGES, FOR EACH VOLUNTEER, IN (a) Az, (b) SURFACE-ZONE (SZ) MINERAL CONTENT, AND (c) LESION-BODY (LB) MINERAL CONTENT FOR TEST AND CONTROL (sucrose-containing gum; n = number of lesions) Test
Subject
n
Control
Change
A B C D E F G H I J N O
11 5 12 14 9 7 11 8 5 4 10 9
8.1 5.9 8.8 4.5 2.9 13.1 4.5 -0.1 14.5 8.9 11.4 5.2
A B C D E F G H I J N O
11 5 12 14 9 7 11 8 5 4 10 9
-1.5 -2.2 2.7 -3.0 -1.9 -3.5 -0.5 -2.9 7.3 -4.8 1.5 0.4
A B C D E F G H I J N 0
11 5 12 14 9 7 11 8 5
10
3.8 3.7 7.9 2.8 -0.1 22.3 0.2 -0.6 11.8 5.3 10.9
9
5.4
4
n
Change
(a) Az (vol.% min. x pm, x 100) 10 2.3 10 2.3 9 7.7 6 7.7 2.8 10 10 12.1 6 1.6 7 -26.8 4 20.0 10 3.8 10 7.3 10 10.5 (b) Surface-zone mineral content (vol.% min.) 10 0.7 10 -0.2 9 2.2 6 -2.2 10 -3.3 10 1.1 6 -5.2 7 6.3 4 -9.6 10 2.1 10 -1.2 10 0.6 (c) Lesion Body (vol.% min.) 10 3.7 10 -0.6 9 3.3 6 -4.0 10 -1.8 10 8.6 6 -1.9 7 -1.0 4 9.3 10 2.4 10 10.0 10 5.6
Difference
95% C.I. for Difference
5.7 3.6 1.1 -3.2 0.1 1.1 2.9 26.7 -5.5 5.1 4.2 -5.3
2.6 -1.6 -2.5 -7.0 -4.5 -4.9 -5.8 14.3 -18.7 -3.8 0.6 -8.7
-2.2 -2.0 0.5 -0.8 1.5 -4.6 4.7 -9.2 17.4 -6.9 2.7 -0.3
-6.6 -5.2 -2.4 -2.9 -5.1 -12.6 -15.3 -23.9
0.2 4.4 4.6 6.8 1.7 13.6 2.1 0.4 2.6 2.9 0.9
-2.5 1.8 1.7 2.0 -2.0 -3.4 -2.2 -0.9 -4.7 -7.5 -2.8
-0.2
-2.8 to 2.5
to to to to to to to to to to to to
8.9 8.7 4.6 0.7 4.6 7.1 11.6 39.0 7.6 13.9 7.7 -1.8
2.2 1.3 3.4 1.4 8.0 3.5 24.9 5.6 54.3 -12.3 to -1.4 0.4 to 5.0 -2.3 to 1.8
to to to to to to to to -19.5 to
to to to to to to to to to to to
2.9 6.9 7.5 11.5 5.5 30.6 6.3 1.7 9.8 13.3 4.6
1899
In situ REMINERALIZATION WITH CHEWING GUM
Vol. 71 No. 12
1 for each volunteer for both the sugar-free gum and the control Analysis of the results for the Az values on an individual volunteer basis, with the 95% confidence intervals in the final column of Table 1, showed that five volunteers (A, B, E, F, J) had a significant difference between the test and control in favor of the sugar-free gum. For one volunteer (D), a significant difference was found in favor of the control, and for the remaining six volunteers the results were inconclusive. For the surface-zone mineral content, the corresponding values were: five in favor of the gum, two in favor of the control, and five inconclusive. For the lesion-body mineral content, the results were: five in favor of the gum, three in favor of the control, and four inconclusive. As a summary of the results, the weighted mean predicted differences between week 7 and week 0 are shown in Table 3 for each parameter. For the chewing gum Az parameter, the value was 788 vol.% min. x pm, corresponding to a remineralization of 18.2% (p < 0.001) vs. baseline average. The control value was 526 vol.% min. x Wm, corresponding to 12.1% remineralization (p < 0.01) vs. baseline average. This difference between 18.2% and 12.1% was not statistically significant (p = 0.07). There were no significant differences between the test and control for the weighted mean surface-zone and lesion-body values (p = 0.20 and p = 0.28, respectively). Sucrose-containing gum.-The mean baseline values for all lesions (n = 207) in the sucrose-containing gum study were: Az, 4066 (vol.% min. x pm); surface zone (SZ), 63.0 (vol.% min.); and lesion body (LB), 42.1 (vol.% min.). The predicted changes in the three parameters are shown in Table 2 for each volunteer for both test and control runs. For the Az values, three volunteers had a significant difference between the test and control in favor of the sucrose-containing gum. For one volunteer, a significant difference was found in favor of the control, and for the remaining eight volunteers the results were inconclusive. For the surfacezone mineral content, one volunteer had a significant difference in favor of the gum, another in favor of the control, and the rest gave inconclusive results. For the lesion-body mineral content, the results were: three in favor of the gum, none in favor of the control, and nine inconclusive. The weighted mean predicted differences between week 7 and week 0 are also summarized in Table 3 for each parameter. For the chewing-gum Az parameter, the value of 743 vol.% min. x pm corresponded to remineralization of 18.3% (p < 0.0001) vs. baseline average. The control value of 438 vol.% min. x pm corresponded to 10.8% remineralization (p < 0.02) vs. baseline average. The runs.
difference between 18.3% and 10.8% was not statistically significant (p 0.08). The weighted mean surface-zone values were not significantly different (p 0.55). However, for remineralization ofthe lesion body, the sucrose-containing gum value of 6.11 vol.% min. was significantly higher (p 0.01) than that of the control (2.81 vol.% min.). =
=
=
Discussion. The aim of this in situ study was to test the hypothesis that chewing gum in a protocol similar to that used in previous experiments (i.e., for 20 min after eating) would result in increased remineralization of artificial enamel lesions beyond that which could be obtained with a fluoridated dentifrice alone. The use of the in situ appliance, initially described by Creanor et al. (1986), links the advantages of single-section specimens with an intra-oral appliance. The analytical method used in this investigation, which allows for variation in baseline lesion size, requires further comment. A more standard method of adjusting for the covariate would be to compute the mean difference (wk7 - wkO) for each subject for each treatment and use the mean baseline size for each subject for each treatment as a covariate in the analyses. This would be a reasonable approach, provided that the relationship between the difference and baseline lesion size is the same for each subject and treatment. The test and control plots in Fig. 2 for a volunteer from the sucrose-containing gum study suggest that the relationships are not the same. This was also assessed by comparison of the fit of parallel lines with the fit of lines with separate slopes. For all three parameters, the model with parallel lines was rejected in favor of the model with individual regression lines, justifying the approach used in the analyses. The protocol used in this experiment entailed the chewing of both the sorbitol- and sucrose-containing gums five times per day, for 20 min, after meals and snacks. As a consequence, any possible deleterious effects of the sugar from the gum should be minimized, coming as they do after the consumption of food. The results indicate that, in both test and control phases of each cross-over study, significant remineralization of the artificial enamel lesions was achieved in these groups of subjects after seven weeks of intra-oral exposure. Clearly, the addition of chewing sucrose-containing gum five times daily for 20 min did not have a detrimental effect on lesion remineralization; indeed, there was a significant difference in favor of the sucrose-
TABL3E 3 SUMMARY OF RESULTS FOR SUGAR-FREE AND SUCROSE-CONTAINING GUM STUDIES 95% C.I. for the Difference in Weighted Mean Predicted Change Diff Weighted Mean Predicted Changes Control Gum Sugar-free Gum Az SZ LB Az
SZ LB
788 2.64 7.55
526 1.65 6.43
743 -0.76 6.11
438
-0.57 2.81
Az - vol.% min. x pm.
SZ - surface-zone mineral content, vol.% min. LB - lesion-body mineral content, vol.% min.
262 0.99 1.12 Sucrose-containing Gum 305 -0.19 3.30
p(1-sided)
-96 to 620 -1.54 to 3.52 -3.04 to 5.28
0.07 0.20 0.28
-145 to 755 -3.60 to 3.22 0.69 to 5.91
0.08 0.55 0.01
1900
CREANOR et al.
containing gum when the mineral content of the lesion body is being considered. These results augment the previous work of Leach et al. (1989). In their in situ study, they reported that salivary stimulation by means of a sorbitol-containing chewing gum not only reversed a trend toward demineralization, but also promoted remineralization. However, the magnitude of the effect of the chewing gum alone is not clear from the protocol used in that study, since the concentration of fluoride in the dentifrice used by subjects was not known. Their in situ model differs from the one reported in this study, where specimens were mounted in a relatively enclosed, protected area, presumably where saliva would reach with greater difficulty. It has been shown (Dawes and Macpherson, 1991) that salivary flow rates are increased during gum chewing to a peak of about ten times the unstimulated flow rate during the first minute of gum chewing followed by a fairly rapid decrease to a plateau at three times the unstimulated flow rate after 20 min. These authors found no difference in salivary flow patterns between chewing gums of different flavors which contained either sucrose or sorbitol. In addition, they found that, for the sucrose-containing gums, the salivary sucrose concentration peaked within the first 1 or 2 min and then fell rapidly. Other studies have examined the effect of chewing gum on plaque pH. Jensen and Wefel (1989) showed that stimulation of salivary flow, with either a sugar-free or sucrose-containing gum, reversed the fall in plaque pH after the consumption of a meal or a snack. That study also reported that there was no statistical difference between the gums in their ability to raise a lowered plaque pH. Park et al. (1990), on the other hand, reported that sucrose-containing gum did not neutralize as much acid as sugar-free gum, particularly when chewing times were under 20 min. Manning and Edgar (1991), however, reported that chewing either sucrose-containing or sugar-free gums for 20 min after an acidogenic challenge increased the rate at which resting plaque pH was reached. It would appear from this present study that the increase in salivary flow that occurs during gum chewing, which is known to accelerate a rise in plaque pH, will not only reduce the tendency to demineralization, but may also enhance the degree of lesion remineralization beyond that obtained with a 1100ppm-F (NaF) dentifrice alone. The significant increase in lesion-body mineral content in the test phase of the sucrose-containing gum study, linked with a tendency toward greater overall lesion mineral content, requires further comment. It is now clear that the ultimate aim of early enamel lesion remineralization should be to direct mineral deposition not only to the surface-layer, but also to the deeper portions of the lesion. In this test, this aim would appear to have been successful, with the only difference in protocol in the test phase being the chewing of sucrose-containing gum five times daily for 20 min. It may well be that the calcium fluoride that will have formed as a result of the dentifrice fluoride exposure (Fejerskov et al., 1981; Featherstone and ten Cate, 1988; Bruun and Givskov, 1991) breaks down in the test phase due to the drop in plaque pH that is sure to exist during exposure to fermentable carbohydrate. The brief additional exposure to an acidic environment coupled with an increased salivary flow (Dawes and
J Dent Res December 1992
Macpherson, 1991), and the consequent rapid increase in plaque pH, may be sufficient to help clear the surface pores ofthe lesion, thereby permitting access of mineral to the deeper portions of the lesion. The trend toward greater remineralization, following the chewing of either sorbitol-or sucrose-containing gum after eating, further substantiates the potential beneficial effects of gumchewing. Extrapolation of these results, however, to any other gum-usage protocol is not justified, and further work is required for investigation of the effects of chewing of either sorbitolcontaining or sucrose-containing gum for periods other than 20 min, and at times other than after meals.
Acknowledgments. We gratefully acknowledge the enthusiastic participation and compliance of all the subjects involved in both studies. REFERENCES
Angmar B, Carlstrom DJ, Glas JE (1963). Studies on the ultrastructure of dental enamel IV. The mineralization of normal human enamel. J Ultrastruct Res 8:12-23. Bruun C, Givskov H (1991). Formation of CaF2 on sound enamel and in caries-like enamel lesions after different forms of fluoride application in uitro. Caries Res 25:96-100. Creanor SL, Strang R, Telfer S, MacDonald I, Smith MJ, Stephen KW (1986). In situ appliance for the investigation of enamel de- and remineralization. Caries Res 20:385-391. Dawes C, Macpherson LMD (1991). Salivary flow rate and sugar clearance from different chewing gums. Caries Res 25:225. Edgar WM, Bibby BG, Mundorff SA, Rowley J (1975). Acid production in plaques after eating snacks: modifying factors in foods. JAm Dent Assoc 90:418-425. Edgar WM, Geddes DAM (1990). Chewing gum and dental health-a review. Br Dent J 168:173-177. Featherstone JDB, ten Cate JM (1988). Physicochemical aspects of fluoride-enamel interactions. In: Ekstrand J, Fejerskov 0, Silverstone LM, editors. Fluoride in dentistry. Copenhagen (Denmark): Munksgaard, 125- 149. Fejerskov 0, Thylstrup A, Larsen MF (1981). Rational use of fluoride in caries prevention. A concept based on possible cariostatic mechanisms. Acta Odontol Scand 39:241-249. Jensen ME, Wefel JS (1989). Human plaque pH responses to meals and the effects of chewing gum. Br Dent J 167:204-208. Leach SA, Lee GTR, Edgar WM (1989). Remineralization of artificial caries-like lesions in human enamel in situ by chewing sorbitol gum. J Dent Res 68:1064-1068. Manning RH, Edgar WM (1991). Effects of chewing gum on plaque pH profiles after a sucrose-containing snack and rinse. Caries Res 25:234. Park KK, Schemehorn BR, Bolton JW, Stookey GK (1990). Effect of sorbitol gum chewing on plaque pH response after ingesting snacks containing predominantly sucrose or starch. Am J Dent 3:185-191. Strang R, Damato FA, Creanor SL, Stephen KW (1987). The effect of baseline lesion mineral loss on in situ remineralization. J Dent Res 66: 1644-1646.
Introduction. The effects of chewing gum use on oral health were reviewed recently by Edgar and Geddes (1990). In evaluations of possible effects of chewing gum on caries, the usage pattern of the gum has to be considered. Edgar et al. (1975), using a five-minute chewing period, investigated the effects of gums alone and found that sorbitol-based chewing gums had little, if any, effect on plaque pH, while sucrose-containing gums lowered plaque pH. Park et al. (1990) investigated the effect of chewing a sorbitol gum for 10 min after the ingestion of snack foods. They found that use of chewing gum resulted in a smaller fall in pH and a quicker rise to resting pH levels. This they attributed to an increase in salivary flow and buffering capacity of the saliva. Other studies have looked at longer chewing periods. Jensen and Wefel (1989) and Manning and Edgar (1991) found that the chewing of sugar-free gums for 20 min after an acidogenic challenge also decreased the time for the resting pH to be regained. Importantly, in both these studies, sucrose-containing gums were found to have the same effect, which were not statistically different from those of the sugar-free gums. There have been few in situ de/remineralization chewing gum studies published. Leach et al. (1989) reported that chewing a sorbitol gum for 20 min after meals and snacks resulted in Received for publication March 13, 1992 Accepted for publication May 27, 1992 This study was supported by a grant from the Wm. Wrigley, Jr. Company (Chicago, USA).
significantly greater remineralization. The aim of this in situ study was to test the hypothesis that chewing either sorbitol- or sucrose-containing gum for 20 min after eating would result in increased remineralization of artificial enamel lesions beyond that which could be obtained with a fluoridated dentifrice alone.
Materials and methods. Two cross-over tests were run separately and were designed to investigate the possibility of enhanced early enamel lesion remineralization with use of chewing gum. The first study was carried out with a sorbitol-containing gum, and the second, which had an identical protocol, tested a sucrose-containing chewing gum. Subjects and experimental protocols.-Twelve healthy adult subjects participated in each study, which had been approved by the local Ethics Committee. Ten subjects were common to both studies. Upper and lower alginate impressions were taken of the subjects' dentitions, from which plaster casts were constructed. Lower appliances, similar to those described by Creanor et al. (1986), were constructed for each subject. These appliances are worn on the lingual aspect of the mandibular arch. Enamel sections are mounted in recessed areas on the fitting surface of the device, positioned just below the lingual aspect of the lower first and second molar teeth (Fig. 1). The volunteers were provided with the devices (without specimens at this point) prior to being tested, to be worn for a few days so that any irritating areas could be adjusted. All subjects were provided with an ample supply of 1100-ppm-F (NaF) dentifrice at least four weeks before the studies. This dentifrice was to be used throughout the seven-week period of both studies, and even when the appliance was not being worn (i.e., between the tests). Volunteers were instructed to brush twice daily (morning and night) for two min, with a pea-sized amount of toothpaste on a Sensodyne 3.5 toothbrush. Those wishing to brush more frequently were supplied with a non-fluoride dentifrice (1.1 ppm). Volunteers were requested not to rinse at the end of the procedure. During the period of the tests, no other fluoride supplement was to be used. Appliances were worn at all times, including during sleeping and eating, apart from a short period to allow for the cleansing of the lingual aspects of the lower natural dentitions. In an attempt to standardize the diet in both studies, we recorded a weekly record ofeach subject's food intake for the first week of each test. A copy was made and the sheet returned to each subject, with the request that he/she should keep to a similar diet throughout both test and control phases of each study. Volunteers were provided with a supply of snack foods, one of which was to be consumed each day at mid-morning and at mid-afternoon as a snack. No other dietary restrictions were imposed. During the first phase of each ofthe studies, six subjects were instructed to chew five sticks of gum per day, each for 20 min, immediately after breakfast, lunch, and dinner, and after each of the snacks. The remaining six subjects acted as controls, but followed the gum protocol during the cross-over phase, while the first six carried out the control portion. For both tests, the gum was supplied by the Wm. Wrigley, Jr. Co. (USA), and for the first 1895
CREANOR
1896
et
J Dent Res December 1992
al.,
Az difference
vol
min
pm)
x
-
2200
2000'
+4
oniroli
Team
1800
16001400-
12001000
P -t
800
600400200 Z%
!
0
"t 4
-200
0
Fig. 1--The in situ appliance, showing the recessed trough areas on the fitting surfface of the device, positioned just opposite the lower first and second molars. Two enamel specimens are mounted on the base of each
troughl
US), This product weighs 2.7 gper stick, sweetener, along with trace amounts of hydrogenated glucose syrup and aspartame The gum for the test
was
Orbit (Extra
with sorbitol
second test
in
the main
as
was
Doublemint,
which contains 2.5 g of
(sucrose, glucose, glucose syrup)
per
Sectionm preparation.-Artificial enamel lesions were created in sound human premolars extracted for orthodontic reasons. Teeth were coated with an acid-resistant nail polish
on
all aspects, apart from four to six windows (each 1 x 10 their buccal Each tooth was immersed for five
in
approximately
mmol/L
ing 2.0
a
demineralization solution
phosphate, and at a pH of ppmn fluoride and was changed
daily. a
12 longitudinal sections were cut 'rom
Microslice 11 Hard Tissue Diamond Saw to
a
each
tooth
thickness
approximately 250 pm; these were hand-ground, with a alumina slurry, to a final measured thickness of between
of
and 140 pm.
(As).
Baseline mineral content measurements
source-specimen distance of 300
mm.
Mineral
al., 1987). Measurements were made of the total mineral loss (As) and the volume percent mineral contents of the surface zone and lesion body. The differences (week 7 week in As values
0)
sitometric
microden multiplied by that, parameters, negative values represent demmneraliza-
tion and
one so
total min-
that thc
amount of
for all
positive values remnineralization within the lesion. -For each lesion, the difference between
Statistical analysis
the baseline value (week 0) and the final value (week 7)
was
such variations, plots of the differences against riade for the lesions from each volunteer
control phases.
These
plots
Regression lines through ifhese plots of thie predicted change
nation by
were
made for As,
and LB.
use
of
baseline
common
a
value,
the
determination
Pt
baseline
nean
of all
of predicted
the test and control
for volunteer
control (P
profiles through the lesions were calculated from a microdensitometric analysis of the microradiographs tAngmar et al., 1963). Microdensitometry was performed with a Leitz ASBA image analyzer connected to a microcomputer (Strang et
minus
microdensitometric parameters: SZ), and lesiorn-body (LB)
were
for the test and
120
were
content
have all been
three
e
shown (Strang et at. 1987)
values
for the test
Microradiograplhy
a
th
(As), surfaicesone
To allow for
baseline
fine
made, and the sections were then varnished on all cut aspects apart from their natural enamel surfaces, At the end of the seven-week period, the sections were detached from the appliances, the varnish removed, and microradiographs repeated. Four sections were allocated to every appliance on each of the test and control phases for both studies. niicrodensitonmetry. -Mi croradiographs of the enamel sections and an aluminum stepwedge were taken, at baseline and at the end of the seven-week experimental period, with a 20-minute exposure on Kodak High Resolution plates (Type IA) by a Cu(Ka-) x-ray source operating at 20 kV and 30 mA at
for
contents. It has been
for
From ten to with
9000
eral
days
contamn-
calcium, 2.0 mmnol/L
4.5. The solution contained 0.01
calculated
mineral loss
mm) SZ,
surfaces.
50 mL of
8000
7000
remineralisation is related to baseline lesion demineralization
in vitro
on
.3000 4000 6000 6000
2000
Baseline Az (% vol minx pm) Fig. 2 Plot of Az diffrence [wkl w kO) x -11 against baseline Az fPr each lesinacr volunteer (A) fnr both the suicrose-containing gum and control rurns. Pt predicted test (gum) and P. predicted control values for thtvolunteer
sugars
3.2-g stick.
1000
A, Fig.
phases.
were
used for determi"
(wk7 -wk0) for each volunteer, value (see Fig. 2). This same lesiorns
in
each study,
was
used
charges for all volunteers for both, Thus fbr sucrose-containing gum
2 indicates that his
predicted change
(Pt) is approximately 810 vol.% min
)is approximately 230 vol.% min.
x
x
pmn and
ptm
in As
for the
These values
reported in the first row of Table 2. A weighted analysis of variance was undertaken with use of these predicted changes. Weighted analysis was used since these predicted changes are based on different numbers of lesions and are, therefore, estimated with differing degrees of precision for each subject and treatment. It was decided that the predicted changes should be weighted by use of the square root of the number of lesions for each subject/treatment. This analysis Was performed with tihe MINITAB statistical package with the general linear models command (GLM). In addition, since of
and
there
was
P.
then
are
some evidence
in both studies of
a
greater diffrence
in favor of the test (i e.,
who used the
gum in the first
was
in the
gum) for those volunteers experimental run, a period effct
included
analyses.
Since
prexvious
difference would
work with gums
be
would
suggest
that ary
in favor of the gum, one-tailed tests
were
used.
Results.
Sigarrfre0gu.The mean baseline values for all lesions (n= 264) in the sugar-free gum study (ie. both gum and control lesions) were: Az, 4332 (vol/v min x jim); surface zone (SZ), 62.0 (vol.% min.); and lesior body (LB) 42.1 (voA min.). The predicted changes in the three parameters are shown in Table
Vol. 71 No. 12
In situ REMINERALIZATION WITH CHEWING GUM
1897
TABLE 1 PREDICTED CHANGES, FOR EACH VOLUNTEER, IN (a) Az, (b) SURFACE-ZONE (SZ) MINERAL CONTENT, AND (c) LESIONBODY (LB) MINERAL CONTENT FOR TEST AND CONTROL (sugar-free gum; n = number of lesions) Test
Subject
n
Control
Change
A B C D E F G H I J K L
16 16 16 11 12 12 16 6 8 12 9 3
15.6 14.9 10.9 1.1 2.5 14.4 4.9 0.2 5.0 4.1 9.0 4.3
A B C D E F G H I J K L
16 16 16 11 12 12 16 6 8 12 9 3
5.7 4.7 5.7 0.7 1.3 7.7 0.2 5.7 1.6 0.8 1.6 -3.8
A B C D E
16 16 16 11 12 12 16 6 8 12 9
12.9 11.8 10.8 0.5 2.0 20.9 4.3 0.1 6.4 4.0 8.6
3
10.8
F
G H I J K L
Change (a) Az (vol.% min. x pm, x 100) 14 4.2 10 3.3 13 7.1 18 6.7 17 -4.1 5 5.4 9 5.0 9 12.8 6 7.2 6 0.2 16 10.8 n
12.1 (b) Surface-zone mineral content (vol.% min.) 14 1.3 10 -2.7 13 2.5 18 2.4 17 -2.0 5 1.5 9 4.2 9 5.2 6 2.3 6 -5.0 16 4.9 4 3.5 (c) Lesion Body (vol.% min.) 14 3.2 3.8 10 5.1 13 9.1 18 2.3 17 5 4.2 9 2.1 9.5 9 10.1 6 1.8 6 10.4 16 4
4
9.8
Difference
95% C.I. for Difference
11.5 11.6 3.7 -5.6 6.6 9.0 -0.2 -12.6 -2.2 3.9 -1.8 -7.7
8.7 8.2 -0.8 -9.1 2.7 1.3 -2.6 -28.2 -7.2 0.1 -5.3 -35.2
to to to to to to to to to to to to
14.2 15.1 8.2 -2.0 10.6 16.6 2.3 3.0 2.8 7.8 1.8 19.7
4.4 7.4 3.2 -1.7 3.3 6.2 -4.1 0.5 -0.8 5.8 -3.3 -7.4
2.8 3.3 1.1 -4.3 0.9 1.3 -7.9 -12.1 -2.9 -0.1 -6.2 -57.2
to to to to to to to to to to to to
6.0 11.5 5.3 1.0 5.7 11.1 -0.2 13.2 1.4 11.8 -0.3 42.5
9.8 8.0 5.8 -8.7 -0.3 16.7 2.3 -9.4 -3.7 2.2 -1.8
6.4 3.3 2.4 -11.7 -3.2 13.0 0.5 -14.6 -6.8 -1.0 -4.1
to to to to to to to to to to to
13.1 12.7 9.1 -5.6 2.6
1.0
20.4
4.0 -4.2 -0.7 5.3 0.6
-8.0 to 10.1
1898
J Dent Res December 1992
CREANOR et al.
TABLE 2 PREDICTED CHANGES, FOR EACH VOLUNTEER, IN (a) Az, (b) SURFACE-ZONE (SZ) MINERAL CONTENT, AND (c) LESION-BODY (LB) MINERAL CONTENT FOR TEST AND CONTROL (sucrose-containing gum; n = number of lesions) Test
Subject
n
Control
Change
A B C D E F G H I J N O
11 5 12 14 9 7 11 8 5 4 10 9
8.1 5.9 8.8 4.5 2.9 13.1 4.5 -0.1 14.5 8.9 11.4 5.2
A B C D E F G H I J N O
11 5 12 14 9 7 11 8 5 4 10 9
-1.5 -2.2 2.7 -3.0 -1.9 -3.5 -0.5 -2.9 7.3 -4.8 1.5 0.4
A B C D E F G H I J N 0
11 5 12 14 9 7 11 8 5
10
3.8 3.7 7.9 2.8 -0.1 22.3 0.2 -0.6 11.8 5.3 10.9
9
5.4
4
n
Change
(a) Az (vol.% min. x pm, x 100) 10 2.3 10 2.3 9 7.7 6 7.7 2.8 10 10 12.1 6 1.6 7 -26.8 4 20.0 10 3.8 10 7.3 10 10.5 (b) Surface-zone mineral content (vol.% min.) 10 0.7 10 -0.2 9 2.2 6 -2.2 10 -3.3 10 1.1 6 -5.2 7 6.3 4 -9.6 10 2.1 10 -1.2 10 0.6 (c) Lesion Body (vol.% min.) 10 3.7 10 -0.6 9 3.3 6 -4.0 10 -1.8 10 8.6 6 -1.9 7 -1.0 4 9.3 10 2.4 10 10.0 10 5.6
Difference
95% C.I. for Difference
5.7 3.6 1.1 -3.2 0.1 1.1 2.9 26.7 -5.5 5.1 4.2 -5.3
2.6 -1.6 -2.5 -7.0 -4.5 -4.9 -5.8 14.3 -18.7 -3.8 0.6 -8.7
-2.2 -2.0 0.5 -0.8 1.5 -4.6 4.7 -9.2 17.4 -6.9 2.7 -0.3
-6.6 -5.2 -2.4 -2.9 -5.1 -12.6 -15.3 -23.9
0.2 4.4 4.6 6.8 1.7 13.6 2.1 0.4 2.6 2.9 0.9
-2.5 1.8 1.7 2.0 -2.0 -3.4 -2.2 -0.9 -4.7 -7.5 -2.8
-0.2
-2.8 to 2.5
to to to to to to to to to to to to
8.9 8.7 4.6 0.7 4.6 7.1 11.6 39.0 7.6 13.9 7.7 -1.8
2.2 1.3 3.4 1.4 8.0 3.5 24.9 5.6 54.3 -12.3 to -1.4 0.4 to 5.0 -2.3 to 1.8
to to to to to to to to -19.5 to
to to to to to to to to to to to
2.9 6.9 7.5 11.5 5.5 30.6 6.3 1.7 9.8 13.3 4.6
1899
In situ REMINERALIZATION WITH CHEWING GUM
Vol. 71 No. 12
1 for each volunteer for both the sugar-free gum and the control Analysis of the results for the Az values on an individual volunteer basis, with the 95% confidence intervals in the final column of Table 1, showed that five volunteers (A, B, E, F, J) had a significant difference between the test and control in favor of the sugar-free gum. For one volunteer (D), a significant difference was found in favor of the control, and for the remaining six volunteers the results were inconclusive. For the surface-zone mineral content, the corresponding values were: five in favor of the gum, two in favor of the control, and five inconclusive. For the lesion-body mineral content, the results were: five in favor of the gum, three in favor of the control, and four inconclusive. As a summary of the results, the weighted mean predicted differences between week 7 and week 0 are shown in Table 3 for each parameter. For the chewing gum Az parameter, the value was 788 vol.% min. x pm, corresponding to a remineralization of 18.2% (p < 0.001) vs. baseline average. The control value was 526 vol.% min. x Wm, corresponding to 12.1% remineralization (p < 0.01) vs. baseline average. This difference between 18.2% and 12.1% was not statistically significant (p = 0.07). There were no significant differences between the test and control for the weighted mean surface-zone and lesion-body values (p = 0.20 and p = 0.28, respectively). Sucrose-containing gum.-The mean baseline values for all lesions (n = 207) in the sucrose-containing gum study were: Az, 4066 (vol.% min. x pm); surface zone (SZ), 63.0 (vol.% min.); and lesion body (LB), 42.1 (vol.% min.). The predicted changes in the three parameters are shown in Table 2 for each volunteer for both test and control runs. For the Az values, three volunteers had a significant difference between the test and control in favor of the sucrose-containing gum. For one volunteer, a significant difference was found in favor of the control, and for the remaining eight volunteers the results were inconclusive. For the surfacezone mineral content, one volunteer had a significant difference in favor of the gum, another in favor of the control, and the rest gave inconclusive results. For the lesion-body mineral content, the results were: three in favor of the gum, none in favor of the control, and nine inconclusive. The weighted mean predicted differences between week 7 and week 0 are also summarized in Table 3 for each parameter. For the chewing-gum Az parameter, the value of 743 vol.% min. x pm corresponded to remineralization of 18.3% (p < 0.0001) vs. baseline average. The control value of 438 vol.% min. x pm corresponded to 10.8% remineralization (p < 0.02) vs. baseline average. The runs.
difference between 18.3% and 10.8% was not statistically significant (p 0.08). The weighted mean surface-zone values were not significantly different (p 0.55). However, for remineralization ofthe lesion body, the sucrose-containing gum value of 6.11 vol.% min. was significantly higher (p 0.01) than that of the control (2.81 vol.% min.). =
=
=
Discussion. The aim of this in situ study was to test the hypothesis that chewing gum in a protocol similar to that used in previous experiments (i.e., for 20 min after eating) would result in increased remineralization of artificial enamel lesions beyond that which could be obtained with a fluoridated dentifrice alone. The use of the in situ appliance, initially described by Creanor et al. (1986), links the advantages of single-section specimens with an intra-oral appliance. The analytical method used in this investigation, which allows for variation in baseline lesion size, requires further comment. A more standard method of adjusting for the covariate would be to compute the mean difference (wk7 - wkO) for each subject for each treatment and use the mean baseline size for each subject for each treatment as a covariate in the analyses. This would be a reasonable approach, provided that the relationship between the difference and baseline lesion size is the same for each subject and treatment. The test and control plots in Fig. 2 for a volunteer from the sucrose-containing gum study suggest that the relationships are not the same. This was also assessed by comparison of the fit of parallel lines with the fit of lines with separate slopes. For all three parameters, the model with parallel lines was rejected in favor of the model with individual regression lines, justifying the approach used in the analyses. The protocol used in this experiment entailed the chewing of both the sorbitol- and sucrose-containing gums five times per day, for 20 min, after meals and snacks. As a consequence, any possible deleterious effects of the sugar from the gum should be minimized, coming as they do after the consumption of food. The results indicate that, in both test and control phases of each cross-over study, significant remineralization of the artificial enamel lesions was achieved in these groups of subjects after seven weeks of intra-oral exposure. Clearly, the addition of chewing sucrose-containing gum five times daily for 20 min did not have a detrimental effect on lesion remineralization; indeed, there was a significant difference in favor of the sucrose-
TABL3E 3 SUMMARY OF RESULTS FOR SUGAR-FREE AND SUCROSE-CONTAINING GUM STUDIES 95% C.I. for the Difference in Weighted Mean Predicted Change Diff Weighted Mean Predicted Changes Control Gum Sugar-free Gum Az SZ LB Az
SZ LB
788 2.64 7.55
526 1.65 6.43
743 -0.76 6.11
438
-0.57 2.81
Az - vol.% min. x pm.
SZ - surface-zone mineral content, vol.% min. LB - lesion-body mineral content, vol.% min.
262 0.99 1.12 Sucrose-containing Gum 305 -0.19 3.30
p(1-sided)
-96 to 620 -1.54 to 3.52 -3.04 to 5.28
0.07 0.20 0.28
-145 to 755 -3.60 to 3.22 0.69 to 5.91
0.08 0.55 0.01
1900
CREANOR et al.
containing gum when the mineral content of the lesion body is being considered. These results augment the previous work of Leach et al. (1989). In their in situ study, they reported that salivary stimulation by means of a sorbitol-containing chewing gum not only reversed a trend toward demineralization, but also promoted remineralization. However, the magnitude of the effect of the chewing gum alone is not clear from the protocol used in that study, since the concentration of fluoride in the dentifrice used by subjects was not known. Their in situ model differs from the one reported in this study, where specimens were mounted in a relatively enclosed, protected area, presumably where saliva would reach with greater difficulty. It has been shown (Dawes and Macpherson, 1991) that salivary flow rates are increased during gum chewing to a peak of about ten times the unstimulated flow rate during the first minute of gum chewing followed by a fairly rapid decrease to a plateau at three times the unstimulated flow rate after 20 min. These authors found no difference in salivary flow patterns between chewing gums of different flavors which contained either sucrose or sorbitol. In addition, they found that, for the sucrose-containing gums, the salivary sucrose concentration peaked within the first 1 or 2 min and then fell rapidly. Other studies have examined the effect of chewing gum on plaque pH. Jensen and Wefel (1989) showed that stimulation of salivary flow, with either a sugar-free or sucrose-containing gum, reversed the fall in plaque pH after the consumption of a meal or a snack. That study also reported that there was no statistical difference between the gums in their ability to raise a lowered plaque pH. Park et al. (1990), on the other hand, reported that sucrose-containing gum did not neutralize as much acid as sugar-free gum, particularly when chewing times were under 20 min. Manning and Edgar (1991), however, reported that chewing either sucrose-containing or sugar-free gums for 20 min after an acidogenic challenge increased the rate at which resting plaque pH was reached. It would appear from this present study that the increase in salivary flow that occurs during gum chewing, which is known to accelerate a rise in plaque pH, will not only reduce the tendency to demineralization, but may also enhance the degree of lesion remineralization beyond that obtained with a 1100ppm-F (NaF) dentifrice alone. The significant increase in lesion-body mineral content in the test phase of the sucrose-containing gum study, linked with a tendency toward greater overall lesion mineral content, requires further comment. It is now clear that the ultimate aim of early enamel lesion remineralization should be to direct mineral deposition not only to the surface-layer, but also to the deeper portions of the lesion. In this test, this aim would appear to have been successful, with the only difference in protocol in the test phase being the chewing of sucrose-containing gum five times daily for 20 min. It may well be that the calcium fluoride that will have formed as a result of the dentifrice fluoride exposure (Fejerskov et al., 1981; Featherstone and ten Cate, 1988; Bruun and Givskov, 1991) breaks down in the test phase due to the drop in plaque pH that is sure to exist during exposure to fermentable carbohydrate. The brief additional exposure to an acidic environment coupled with an increased salivary flow (Dawes and
J Dent Res December 1992
Macpherson, 1991), and the consequent rapid increase in plaque pH, may be sufficient to help clear the surface pores ofthe lesion, thereby permitting access of mineral to the deeper portions of the lesion. The trend toward greater remineralization, following the chewing of either sorbitol-or sucrose-containing gum after eating, further substantiates the potential beneficial effects of gumchewing. Extrapolation of these results, however, to any other gum-usage protocol is not justified, and further work is required for investigation of the effects of chewing of either sorbitolcontaining or sucrose-containing gum for periods other than 20 min, and at times other than after meals.
Acknowledgments. We gratefully acknowledge the enthusiastic participation and compliance of all the subjects involved in both studies. REFERENCES
Angmar B, Carlstrom DJ, Glas JE (1963). Studies on the ultrastructure of dental enamel IV. The mineralization of normal human enamel. J Ultrastruct Res 8:12-23. Bruun C, Givskov H (1991). Formation of CaF2 on sound enamel and in caries-like enamel lesions after different forms of fluoride application in uitro. Caries Res 25:96-100. Creanor SL, Strang R, Telfer S, MacDonald I, Smith MJ, Stephen KW (1986). In situ appliance for the investigation of enamel de- and remineralization. Caries Res 20:385-391. Dawes C, Macpherson LMD (1991). Salivary flow rate and sugar clearance from different chewing gums. Caries Res 25:225. Edgar WM, Bibby BG, Mundorff SA, Rowley J (1975). Acid production in plaques after eating snacks: modifying factors in foods. JAm Dent Assoc 90:418-425. Edgar WM, Geddes DAM (1990). Chewing gum and dental health-a review. Br Dent J 168:173-177. Featherstone JDB, ten Cate JM (1988). Physicochemical aspects of fluoride-enamel interactions. In: Ekstrand J, Fejerskov 0, Silverstone LM, editors. Fluoride in dentistry. Copenhagen (Denmark): Munksgaard, 125- 149. Fejerskov 0, Thylstrup A, Larsen MF (1981). Rational use of fluoride in caries prevention. A concept based on possible cariostatic mechanisms. Acta Odontol Scand 39:241-249. Jensen ME, Wefel JS (1989). Human plaque pH responses to meals and the effects of chewing gum. Br Dent J 167:204-208. Leach SA, Lee GTR, Edgar WM (1989). Remineralization of artificial caries-like lesions in human enamel in situ by chewing sorbitol gum. J Dent Res 68:1064-1068. Manning RH, Edgar WM (1991). Effects of chewing gum on plaque pH profiles after a sucrose-containing snack and rinse. Caries Res 25:234. Park KK, Schemehorn BR, Bolton JW, Stookey GK (1990). Effect of sorbitol gum chewing on plaque pH response after ingesting snacks containing predominantly sucrose or starch. Am J Dent 3:185-191. Strang R, Damato FA, Creanor SL, Stephen KW (1987). The effect of baseline lesion mineral loss on in situ remineralization. J Dent Res 66: 1644-1646.
