| Sign In to gain access to subscriptions and/or personal tools. |
Professionally Applied Fluoride Gel in Low-caries 10.5-year-oldsDepartment of Preventive and Restorative Dentistry/117, Radboud University, Nijmegen Medical Center, PO Box 9101, NL-6500 HB Nijmegen, The Netherlands; Correspondence: * corresponding author, g.truin{at}dent.umcn.nl
The question has been raised whether low-caries children regularly using fluoride toothpaste will benefit from the professional application of additional fluoride gel. To investigate the caries-reducing effect of semi-annually-applied neutral 1% sodium fluoride gel, we carried out a double-blind randomized controlled clinical trial (n = 594) in a child population, initially aged 9.5–11.5 years, with baseline caries experience of D3MFS = 0 (decayed, missing, and filled tooth surfaces of permanent teeth). The mean number of tooth surfaces saved from caries development by fluoride gel application after 4 years was 0.2 D3MFS (SE = 0.17). The preventive fraction (PF) showed a mean relative effect of professionally applied fluoride gel of 18%. The cariostatic effect of the fluoride gel on pits and fissures would have been influenced by the sealant strategy in the study. Professionally applied fluoride gel showed no statistically significant effect on mean D3MFS score in low-caries 9.5- to 11.5-year-olds.
Key Words: caries prevention • fluoride gel • randomized controlled trial
A meta-analysis on the caries-inhibiting effect of fluoride gel treatment showed a mean prevented fraction of 22% (van Rijkom et al., 1998). However, the included studies had been carried out among highly caries-active children. Over the last two decades, caries has declined dramatically in most Western countries. This is characterized by a negatively skewed caries distribution among the young. There exists consensus, among experts in the field of preventive dentistry, that caries-free subjects are unlikely to benefit substantially from regular in-office fluoride applications (Gelardi, 1992; Clarkson et al., 1996; Horowitz and Ismail, 1996). However, evidence-based data to support this statement are virtually lacking. Therefore, recommendations regarding the effectiveness of professionally applied fluoride gel in low-caries populations can be based only on extrapolations from results of studies in high-caries populations. The purpose of this randomized controlled trial was to investigate the caries-reducing effect of semi-annual professional sodium fluoride gel application in a low-caries child population over a period of 4 years’ follow-up.
Study Population and Sample The study population consisted of low-caries-risk children, aged 9.5 to 11.5 yrs, who were regular attenders at 3 pediatric dental clinics in the cities of Oss, Nijmegen, and Beuningen in The Netherlands. The drinking water in these cities is not fluoridated (< 0.3 ppm F–). Dental care was performed by 15 dentists and dental nurses. The DMFS index was used to score the number of decayed, missing, and filled tooth surfaces of permanent teeth. ‘Decayed’ referred to caries in dentin (D3). ‘Caries low risk’ was defined as D3MFS = 0 at baseline, based on clinical and radiographic assessment. Eligible children were randomly assigned to the fluoride or the placebo gel group. Gel application was done every 6 mos. The recommended concentration of professionally applied fluoride gel (0.4–0.6%) for children in the Netherlands was used. The protocol of the study was approved by the research ethics committee of Radboud University (CEOM Nr. 9406-0682). A difference of 0.5 D3MFS between the fluoride and the placebo groups after 4 yrs was considered clinically important. The power calculation (80% and a = 5%) resulted in 252 subjects per treatment group. Informed consent was obtained from parents or legal representatives of all participating children.
From the selected age range, we pre-screened a total of 1091 children for caries activity, according to their dental records in 1995. The inclusion criteria were: a mean D3MFS score of 0, and children who regularly attend pediatric dental clinics. The exclusion criterion was absence of informed consent. Forty-two percent of the initially selected subjects refused to participate (n = 457) (Table 1
Clinical and Radiographic Examinations Clinical examinations were carried out according to a written protocol. Caries diagnosis, distinguishing enamel (D2) and dentinal (D3) scores, was carried out according to criteria described by Marthaler (1966), with the use of an operating light, mouth mirror, blunt dental probe, and compressed air. Bitewing radiographs were taken when these were either not available or were older than 1 yr. Bitewing radiographs were available in 80% of the per protocol and in 73% of the intention-to-treat subjects at baseline. Bitewing radiographs were taken in 95% of the per protocol and in 94% of the intention-to-treat subjects at evaluation year 4.
Treatment Interventions were undertaken according to a written protocol. Preventive treatment at the semi-annual check-ups included oral hygiene instruction, followed by supervised toothbrushing with fluoride toothpaste. Subsequently, either a placebo gel or a neutral 1% sodium fluoride gel (4500 ppm fluoride ion) was professionally applied in a flexible disposable tray and retained for 4 min. After the application, the participants were instructed to refrain from rinsing, eating, and drinking for 30 min. The effectiveness of exactly 8 applications was assessed for all subjects who followed the protocol over the 4 yrs. Composite resin sealants were applied in permanent teeth diagnosed with enamel caries lesions (discoloration of fissures) and in caries-free molars in subjects who had a dentin lesion or a restoration elswhere in the permanent dentition.
Statistical Analysis The treatment effect of fluoride gel was tested by multiple regression analysis (ANCOVA) after correction for the initial caries experience (D2S, d2mfs). D2 and d2 refer to enamel caries lesions in the permanent and primary dentition, respectively. The M-condition in permanent teeth was not observed in this study population. The Box-Cox transformation was applied (e.g., double root transformation) because of the skewness of the outcome variables. Due to the low number, analysis of affected buccal and palatal pits and fissures, and buccal and palatal tooth surfaces, had to be restricted to logistic regression. An alpha of 0.05 was chosen for statistical testing. The treatment effect was tested one-sidedly, since a caries-promoting effect of fluoride can be excluded on theoretical grounds. The treatment groups were compared with respect to the percentage of incident cases. A more precise analysis was obtained for the mean D3MFS score, which better reflects the restorative treatment need. The mean D3MFS score is equal to the mean caries increment in the placebo group (Ip) and in the fluoride group (If), since there were no initial caries lesions. The mean D3MFS score for the second molars is interpretable as the risk for caries in the second molars. Therefore, the difference in mean D3MFS score will be called Attributable Risk (AR = Ip-If). The interim analysis (performed in 1999) showed no statistically significant effect on the primary outcome mean D3MFS score (p = 0.73). We used a paired t test to analyze the rate of caries development between baseline and two-year evaluation, on the one hand, and two-and four-year evaluations, on the other. Kappa values expressed the reproducibility for the combined clinical examination and bitewing radiography in the permanent dentition at evaluation year 4, including (a) sound, (b) enamel lesion or sealant, and (c) dentin lesion or restoration. In addition to the principal investigator, 10 observers were involved in the study. The intra-observer agreement of the principal examiner (based on 6753 surfaces) was 0.96, and the inter-observer agreement between the principal and the other examiners varied between 0.90 and 0.98 (based on 850 to 3700 surfaces).
At the start of the study, the mean age of the children was 10.4 yrs (SD = 0.6) in both treatment groups. Baseline caries experience (D2S) in the permanent dentition was 3.8 (SD = 3.0) in the placebo and 3.9 (SD = 3.0) in the fluoride group. The mean number of occlusal sealants in permanent molars at baseline was 3.4 (SD = 2,7) and 3.4 (SD = 2.7) in the placebo and fluoride groups, respectively. In the primary dentition, enamel and dentin caries experience (d2,3mfs) in the placebo and fluoride groups was 3.2 (SD = 4.9) and 3.7 (SD = 5.7), respectively, at baseline. The number of follow-up yrs was 4.0 (SD = 0.1) in both the placebo and the fluoride groups. The exposure time for the second molars in the placebo and fluoride groups was 2.4 (SD = 1.1) and 2.3 (SD = 1.1) years, respectively.
No statistically significant differences in caries increment of the permanent dentition (mean D3MFS score) were found between the placebo and fluoride groups in both the intention-to-treat and the per protocol subjects (Table 2
The AR during 4 yrs was 0.05 (95% CI, 0 to 0.12) per yr. The Preventive Fraction (PF) showed a mean relative effect of professionally applied fluoride gel of 18% in both the permanent dentition and the erupting second molars. Moreover, the mean percentage of incident cases after 4 yrs was 40% in both the placebo and the fluoride groups (Table 3  ).
The difference in mean numbers of sealants in occlusal, buccal, and palatal pits and fissures after 4 yrs was not statistically significant (p = 0.30) in the placebo group (5.6; SD 3.7) compared with the fluoride group (5.2; SD 3.5).
Caries increment (mean D3MFS experience) from baseline to evaluation at year 2 was significantly (p < 0.01) lower than that from evaluation during years 2 to 4 (Fig.
The randomization procedure resulted in two homogeneous treatment groups. Due to the fact that the primary outcome (D3MFS) was similar for intention-to-treat subjects and for per protocol subjects, it was concluded that the low number of non-adherence subjects had no effect on the primary outcome.
The study showed no significant effect on mean D3MFS score of the applied 0.45% fluoride gel. This non-significance is not due to lack of power in the study. A clinically relevant difference of 0.5 D3MFS between the treatment groups could have been demonstrated with a power of 91% when tested one-sidedly at alpha = 0.05 with the observed SD in D3MFS of 1.9 (pooled SD from Table 2 The aim of the trial was to assess the inhibiting effect of professionally applied fluoride gel on dentin lesion development in low-caries children aged 9.5 to 11.5 yrs. It should be noticed, however, that intervention at the semi-annual dental check-ups included a sealant application strategy. In addition, from the age of 5 or 6, children attended the pediatric dental clinics for regular dental care. Preventive treatment included oral hygiene instruction, followed by supervised toothbrushing and fluoride gel application if needed. As a consequence, children in the placebo group probably had a fluoride history. However, it is known that past fluoride use does not necessarily protect the tooth from new caries lesions (Groeneveld et al., 1990). Since 1982, the generally recommended concentration of professionally applied fluoride gel for children in the Netherlands is 0.4–0.6%. The effect of 4500 ppm F was statistically tested one-sidedly, because there is sufficient knowledge to support the statement that the placebo treatment cannot lead to a lower caries incidence. The mean relative effect of professional fluoride gel treatment on D3MFS was 18% in both the total permanent dentition and in the erupting second molars. The present results are in line with results found in fluoride gel studies in populations with a higher caries activity. Van Rijkom et al.(1998) calculated a mean PF of 22% PF (95% CI: 18 to 25%), and Marinho et al.(2003) found a mean PF of 21% PF (95% CI: 14 to 28%). We conclude that professionally applied fluoride gel (4500 ppm fluoride ion) showed no statistically significant effect on D3MFS increment in low-caries-active subjects initially aged 9.5 to 11.5 yrs over a four-year period. Quantitatively, the mean treatment effect was considered not clinically relevant.
The study was supported by a grant (SGZ/16524/95) from the ‘College van zorgverzekeringen’, Amstelveen, The Netherlands. Received for publication June 25, 2003. Revision received December 22, 2004. Accepted for publication February 1, 2005.
Journal of Dental Research, Vol. 84, No. 5,
418-421 (2005) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
