slow-release flouride glass devices: in vivo flouride release and retention of the devices in...

258European Archives of Paediatric Dentistry // 7(4). 2006

Slow-release flouride glass devices: In vivo flouride release and retention of the devices in children

G.A. Andreadis, K.J. Toumba, M.E.J. CurzonDepartment of Paediatric Dentistry, Leeds Dental Institute, Leeds, UK

Key words: Fluoride, slow-release, glass intraoral devices, caries, retention

Postal address: Prof. K.J. Toumba, Dept. Paediatric dentistry, Leeds Dental Institute, Clarendon Way, Leeds, England, LS2 9LU.

Email: k.j.toumba@leeds.ac.uk

AbstractAims: To study the fluoride release and retention of new-

shaped glass slow-release F devices (SRFD) in vivo in chil-

dren. Methods: A group of 15 children had one slow-release

F glass device with 17.4% or 14.8 mgm F with a relative sol-

ubility of 294, attached intraorally on the buccal surface of

one of their maxillary permanent molars. The devices were

monitored for 6 months. Results: The retention rate of the

devices was 93% and the mean salivary F levels in the child

volunteers was 0.17±0.1 ppm F compared with 0.025 ppm F

baseline levels. The difference was always statistically sig-

nificant (p<0.01). There were no systemic or local side-

effects. Conclusions: The devices were well tolerated by

children and released fluoride intraorally for the whole peri-

od of six months to a level consistent with inhibition of dem-

ineralisation.

IntroductionA slow releasing glass device was first used in ruminant ani-

mals, such as cows and sheep, to treat trace element defi-

ciencies of selenium, cobalt and copper, which are very

common in many parts of the world. A variant of this device

was developed for use in humans as a slow releasing fluo-

ride glass device (SFGD) in order to provide longer term

intra-oral fluoride (F) [Toumba and Curzon, 1993; Toumba,

2001; 2004]. This type of device dissolves slowly when

moist with saliva, releasing F into the mouth. Initially the

SFGD device was dome shaped, the outer surface resem-

bling a disk, with a diameter of 4mm. This type of device was

tested in a double-blind, controlled in vivo study and was

proven very effective in diminishing caries in high caries-risk

children. However, the retention of the devices was not sat-

isfactory [Toumba and Curzon, 2005].

Recently, a new-shaped device has been developed to

improve the retention. The SFGD is now kidney shaped,

about 6mm long, 2.5mm high, and 2mm thick, and weighs

almost 0.1 gr. One surface is flat to attach to the buccal sur-

face of the tooth. The opposite, outside, surface is convex,

thus exposing a larger surface to the oral environment. All

around the remaining surfaces of the device a groove is

placed to enhance retention.

Because of the new shape and different type of retention fur-

ther studies were required to see if the new device was well

retained and equally as effective in releasing F. The aims of

the present study were, therefore, to test a new-shaped

glass slow-release F device for its ability to release F in vivo

in children for a period of six months. Also, to investigate the

retention of the devices and any possible side-effects either

local or systemic.

Materials and methodsSample selection. A group of children aged 6-16 years old

who attended the Paediatric Dental Clinic of the Division of

Child Dental Health of Leeds Dental Institute were recruited

for the study. Although this was a pilot study, a sample

power calculation was used to determine the size of the

sample. This was based on a formula using a comparison of

successes. As the main reason for this study was to investi-

gate whether there was an assumed better rate of retention

so a retention rate of 90% was set as an anticipated out-

come. Using the formula:

Success x failure in study A + Success x failure in study B

n = x magic number

(Success on study A – Success on study B)2

The result of this sample power calculation was:

90 x 10 + 48 x 52

n = x 7.8 = 15

(90-48)2

where n was the sample size, the expected outcome on

study A = present study assuming a 90% success and

where the success rate of study B (the study by Toumba and

Curzon, 2005) was 48% and the magic number was 7.8 for

a significance level of 5% with a statistical power of 80%.

Therefore a group of 15 children was recruited.

All the required information was provided to the Research

Ethics Committee of Leeds and ethical approval was

obtained. An information sheet was provided to every child

volunteer and a different, more detailed one to their parents.

Consent was obtained for every child participating in

the study.

Attachment of the glass devices. Each volunteer had one

glass device, weighing approximately 1gm (containing

259European Archives of Paediatric Dentistry // 7(4). 2006

Slow release flouride

17.4% F or 14.8 mgm F). Each device was manufactured to

have a relative solubility of 294 (on a scale where 1,000 is

highly soluble and 0 is virtually insoluble). The devices were

attached to the buccal surface of each child’s maxillary right

first or second permanent molar using the acid etch com-

posite resin technique (Figure 1a, b). The tooth surface was

cleaned with a F-free paste (Zircate®), washed for 10 sec-

onds and air dried with oil-free air for 10 seconds. Then the

tooth surface and the fitting surface of the glass device were

etched for 30 seconds with 37% orthophosphoric acid gel,

and a thin layer of light cure bonding agent (Scotchbond

Universal®, Kerr UK Ltd) was applied to the etched surface

of the tooth, and light cured for 20 seconds. Then

Spectrum® universal shade light curing composite (UK Ltd,

Bretton) was applied to the fitting surface of the device and

to the tooth surface and light cured for 60 seconds. Any

composite resin cuffs were smoothed with Soflex® discs.

Moisture control was achieved with the use of cotton rolls

and high-speed suction. The whole procedure lasted

approximately fifteen minutes.

Child volunteers were not given a non-fluoride toothpaste to

brush. Saliva samples were collected from the children

before attaching the devices to obtain the baseline levels

and at the end of the study, day 180.

Patient Satisfaction – Gingivitis. Any possible effects on soft

tissues (buccal mucosa, gingiva, etc) were recorded.

Gingivitis was assessed on days 1, 90, and 180 according to

the Gingival Index (GI) by Löe, [1967]. Plaque was scored on

the same days using the Plaque Index (PI) [Löe, 1967]. All

soft tissues were checked by visual inspection, on days 1,

90, and 180 for ulcers, erythema or tissue irritation.

A questionnaire was given to all parents of the children upon

completion of the study. They were asked to give their opin-

ion regarding any possible complaint or difficulty that their

children had encountered, or complained of, when wearing

the device.

Retention rates – Durability – Safety of the devices. Any

damaged or lost devices were recorded to assess the reten-

tion rates of the new-shaped devices in the children.

Statistical Analysis. The intraoral F levels at baseline and at

the end of the study, as well as the initial and final scores of

the GI and the PI, were compared with both Student’s t-tests

(paired) and with non-parametric Wilcoxon paired tests.

Both tests agreed on statistical significance; thus, the t-test

was finally used for the in vivo comparisons. Statistical

analysis was performed by the use of a computer software

package (SPSS, version 10.0).

ResultsAssessment of the volunteers. Fifteen fit and healthy children

aged 6 to 16 years old (mean age 9.7±3.7) were recruited; 11

of the 15 were less than 9 years old. The mean (±SD)

DMFT+dmft was 3.55±3.1 and the mean DMFS+dmfs was

4.93±4.4. There were eight girls and seven boys.

Daily fluoride release of the devices. Saliva samples were

collected from the children at the baseline and at the end of

the study (day 180). The mean baseline salivary F level was

0.025±0.005 ppm; at day 180 the mean salivary levels (Table

1) was 0.17±0.10 ppm F (95%CI 0.09-0.21). When compar-

ing the F levels at baseline and at day 180 there was a stat-

istically significant difference (p<0.01) with the paired t-test.

Retention rates. All children managed to attend for review at

the three-month recall. A thorough clinical examination was

performed using a blunt probe and a mirror. All the devices,

except one, were in place. The device that was lost had

been attached to the molar of a girl and she told us that it

was lost about three weeks after attachment, while she was

eating. The device was swallowed and no adverse effects

were reported. The remaining 14 children had their devices

intact, and they reported that they had forgotten about its

Collection of saliva samples. Saliva samples were collected

from the children into plastic specimen containers as whole

mixed un-stimulated saliva before attaching the devices to

obtain the baseline salivary F levels of each individual. Saliva

was pooled for two minutes in the mouth and then collect-

ed. The samples were then analysed using a F ion-specific

electrode. At the follow-up visits at 30 and 180 days saliva

samples were also collected and analysed to determine the

amount of F present. Saliva samples collected from the vol-

unteers, were either analysed immediately, within 12 hours

of collection, or frozen at -12°C until analysis using an ion-

specific electrode (Fluoride combination electrode Orion

920A Ionanalyser Meter) took place.

Figure 1:

Intra-oral photo-

graphs showing:

a) a device

attached to the

maxillary right first

permanent molar

of a 9 year old boy

positioned so as

not to interfere with

the occlusion;

b) attachment of

a glass slow-release

F device on the

right maxillary first

permanent molar

of an 8 year old girl.

260European Archives of Paediatric Dentistry // 7(4). 2006

Andreadis et al.

existence a few days after it was placed. No local side-

effects were observed. The retention rate at three months

was 93%. Children were examined again at the end of 6

months. All 14 devices were in place, retention rate 93%

(Table 1). However, three of the 14 devices were damaged,

as small pieces of the glass had dissolved leaving tiny voids.

Overall, there was no significant change to the initial shape

or size of the devices.

Side-effects: local and systemic. The devices did not cause

any local or systemic side-effects. All soft tissues (buccal

mucosa, gingiva, etc) were healthy with no sign of any irrita-

tion, ulceration or erythema. Gingivitis of the relevant tooth

was assessed on days 1, 90, and 180 according to the

Gingival Index (GI) and the Plaque Index (PI) by Löe [1967].

There was no significant difference when comparing the

indices at day 1, and at days 90 and 180, using the t-test

(p>0.05) (Table 1). There was only a tendency for increased

plaque retention on the top of the devices. The gingiva on

the buccal surface of the teeth was also the worst site for

severity of bleeding.

DiscussionAll fifteen children mentioned that they felt the device com-

fortable all the time and no child felt any soreness. There

were 12 out of the 15 children who reported that they got

used to the device in 2-3 days. The rest got used to their

device in 4-7 days. Overall, all the volunteers and their par-

ents stated that they were very satisfied with the devices.

Retention rates at six-month follow-up. The retention rate at

six months was 93% for the children. That was a very

encouraging outcome. The difference was statistically signif-

icant when compared with the 48% retention rate that was

achieved during the only previous similar study [Toumba and

Curzon, 2005]. The new design of the devices (kidney shape

with a circumferential retentive groove) proved quite effec-

tive. In addition, it has to be mentioned that quite a large

amount of composite resin was used to attach each device,

which probably provided a substantial bulk able to with-

stand forces from mastication and tooth brushing.

The devices were attached to the molars using the acid etch

composite resin technique which proved satisfactory.

However, there was a disadvantage to this technique. The

devices are expected to release F for 2-3 years in children as

the aim of developing this preventive device was that it

could be in place for up to two years. This would have great

advantages in groups of low-socioeconomic high caries risk

children where attendance is known to be poor. After this

period, the whole procedure has to be carried out again to

re-attach the new devices but this is an advantage as it

enables the effect of the device to be monitored.

A development of a type of bracket to hold the device in

place and allow for easy reloading, would be ideal. In addi-

tion, removing the bulk of the composite resin at the end of

the study in order to remove the devices was a difficult and

demanding task in order to avoid tooth damage.

Attachment of the F device. All the children had the device

attached on one of their maxillary first permanent molars.

The attachment process was carried out without any prob-

lems, however, the procedure was more difficult in younger

children as this tooth had not erupted completely and con-

sequently there was not enough crown height available to

easily accommodate the device.

One slow-release F device including 17.4% F and with a sol-

ubility of 294 was placed to each of the volunteers. One of

the upper first permanent molars was selected for attaching

the device as Toumba (1996) has shown that this was the

most comfortable and effective site in the mouth.

Attachment to the buccal surface of a maxillary primary sec-

ond molar should be possible. However, smaller devices

might be needed.

Daily Fluoride release of the devices in vivo. Shields et al.

[1987] found that subjects with no caries experience, from

both fluoridated and non-fluoridated areas, had salivary F

levels of 0.04 ppm or greater, whereas high caries patients

from both fluoridated and non-fluoridated communities had

salivary F levels of 0.02 ppm or less. This was also true for

salivary F levels related to caries in the primary dentition

[Toumba and Curzon, 2001]. In our study the baseline mean

Table 1 Gingival, Plaque Indicies (mean scores) and mean ± standard deviation fluoride saliva concentrations in a study using child volun-

teers wearing a slow fluoride releasing glass device.

Day Plaque Index Mean score Retention of devices Fluoride* conc.

Gingival Index

Baseline 1.78 0.55 15/15 0.025±0.05

90 1.88 0.69 14/15 NR

180 1.94 0.65 14/15 0.17±0.01

* = mgm/L; NR = not recorded

261European Archives of Paediatric Dentistry // 7(4). 2006

Slow release flouride

salivary F levels for children was 0.025 ppmF and at day 180

the mean salivary levels were 0.17±0.1 ppm F. Thus, the

glass SRFD devices provided elevated F levels, well above

0.04 ppm.

In another study by Cain et al. [1994], it was shown that a

different type of SRFD could achieve dose-related reminer-

alisation of enamel lesions in situ. This study showed clear-

ly that the constant elevation of salivary F produced signifi-

cant remineralisation which was effective and dose related.

There was a three-fold salivary F elevation and a similar

increase in percent remineralisation efficacy.

Finally, in an in vivo study by Toumba [1996], the salivary F

levels for a test group of children, wearing glass devices with

13.3% F, was 0.17 ppm after attachment and 0.11 ppm at

completion of the study after two years.

It is clear that the achieved F levels of about 0.17 ppmF,

were well above the levels that other studies have proven as

efficient in achieving remineralisation. This indicates the clin-

ical significance of the result. We can assume that the ele-

vated F levels achieved at the present study, would have

inhibited demineralisation and promoted remineralisation for

the whole period of six months. Further randomised long

term controlled clinical studies are now needed in order to

prove the efficacy of the devices to prevent caries and alle-

viate dentine sensitivity.

Side-effects – Patient comfort. The devices did not seem to

cause any discomfort or any local and/or systemic adverse

effects to the children. Increased gingivitis was expected at

the site of attachment of the device, but it was interesting to

notice that the gingival index (GI) and the plaque index (PI)

did not deteriorate significantly. As some children men-

tioned, the device acted for them as a reminder to brush

their teeth better and longer. However, there was elevated

plaque retention on the top, gingival margin, of the devices.

This was prominent in all of the younger children, as the first

permanent molars were not fully erupted and there was

not enough space for attaching the devices to be clear of

the gingiva.

There were 12/15 children who reported that they got used

to the device in 2-3 days. The rest got used to their device

in 4-7 days. It seems that, although after attaching the

device all the volunteers felt there was a big bulk attached to

their maxillary molar, and they were tempted to try to remove

it, all of them reported that after a few days they forgot about

its existence.

ConclusionsThe mean salivary F levels achieved in vivo with the 17.4%

devices (solubility 294) in children was 0.17 after 6 months,

compared with 0.025 ppm at baseline. The retention rates

of the devices were 93% (14/15) in children and the new

shape of the devices improved their retention enormously.

ReferencesCurzon MEJ, Toumba KJ. In vitro and in vivo assessment of a glass slow flu-

oride releasing device. Br Dent J. 2004:196:37-46.

Cain BE. Corpron RE. Fee CL. Strachan DS. Kowalski CJ. Dose related rem-

ineralization using intraoral fluoride-releasing devices in situ. Caries Res

1994;28(4):284-90.

Löe H. The Gingival Index, the Plaque Index and the Retention Index Systems

J Periodontol 1967;38: 610-6.

Shields C. Leverett D. Adair S. Featherstone J. Salivary fluoride levels in fluor-

idated and non-fluoridated communities. J Dent Res Special Issue

1987:141, Abstr 277.

Toumba Kj, Curzon MEJ. Slow release flouride. Caries Res 1993; 27 (suppl 1):

43-46.

Toumba KJ. In vivo and in vitro evaluation of a slow-release fluoride glass for

the prevention of dental caries in high-risk children Ph.D. thesis, Leeds

Dental Institute, 1996

Toumba KJ. Slow-Release Devices for Fluoride Delivery to High-Risk

Individuals. Caries Res 2001;35 (suppl 1) : 10-13, 2001.

Toumba KJ. Slow release devices for flouride delivery to high risk individuals.

Caries Res 2001;35 (suppl 1) : 10-13.

Toumba KJ, Curzon MEJ. Clinical trial of a slow releasing glass fluoride device

to prevent dental caries in children. Caries Res 2005;39:201-206.

Toumba KJ, Curzon MEJ. Flouride concentrations in saliva related to caries in

primary teeth. Europ J. Paediatr Dent 2001; 2:15-19