- Email: [email protected]
An approach to differentiate between antibacterial and antiadhesive effects of mouthrinses in vivo
ARCHIVES OF
PERGAMON
Archives of Oral Biology 43 (1998) 559±565
ORAL BIOLOGY
An approach to dierentiate between antibacterial and antiadhesive eects of mouthrinses in vivo R. Weiger *, L. Netuschil, T. Wester-Ebbinghaus, M. Brecx Department of Conservative Dentistry, School of Dental Medicine, University of TuÈbingen, Osianderstr. 2±8, 72076 TuÈbingen, Germany Accepted 3 March 1998
Abstract An experimental set-up allowing dierentiation in vivo between antibacterial and antiadhesive properties of mouthrinses is described. The percentage of vital bacteria (=microbial vitality) and the bacterial counts were microscopically evaluated in saliva and in supragingival dental plaque both collected simultaneously at various times during de novo plaque formation. In a cross-over design, 12 healthy participants refrained from all oral hygiene for four separate periods of 2 4 h and 2 72 h after having rinsed with either an amine ¯uoride/stannous ¯uoride solution (Meridol1) or 0.9% NaCl (placebo). Stimulated whole saliva was collected before and after the rinse. Together with whole-saliva samples, representative 4, 24 and 72-h-old plaque samples were separately taken from de®ned vestibular tooth surfaces that had been either exposed to the mouthrinse (unprotected sites) or temporarily covered with inert plastic ®lms (protected sites) during rinsing. The pooled plaque and saliva were stained with ¯uorescent dyes to dierentiate vital from dead micro-organisms which permitted the estimation of the percentages of vital bacteria. The total bacterial counts were quanti®ed under the dark®eld microscope. The Wilcoxon test was used for selected pairwise comparisons (a = 0.05). The percentage of vital bacteria in saliva fell signi®cantly from 80±95% to about 50±60% as a result of the antibacterial activity of the test solution. These baseline values and those found in the presence of 4 and 24-h-old plaque were frequently lower than those recorded after the placebo rinse. In comparison to the placebo, microbial vitality was signi®cantly reduced in early supragingival plaque formed on unprotected sites after applying the test solution. The similar total bacterial counts in 4-h-old plaque recorded after the use of the test solution on the unprotected and the protected areas did not point to an antiadhesive eect of the agent. It is concluded that this new experimental set-up allows decoding of the mode of action of a mouthrinse. # 1998 Elsevier Science Ltd. All rights reserved. Keywords: Mouthrinse; Dental plaque; Saliva; Amine ¯uoride; Stannous ¯uoride
1. Introduction Mouthrinses used daily should inhibit or suppress supragingival dental plaque formation and thereby prevent caries and gingivitis. A variety of dierent oral chemotherapeutics are now available, with chlorhexidine being the most ecient antiplaque agent (Adams
* Corresponding (49)7071-295656
author.
Tel.:
(49)7071-2983498;
fax:
and Addy, 1994; Brecx, 1997). Most of them have antibacterial properties, which explain their eect on plaque development. From a biological point of view, however, it is desirable to maintain the natural balance of the oral micro¯ora rather than to disturb it by the use of any antibacterial mouthrinse. For this reason, substances with no or only slight antimicrobial activity are sought. A ®rst approach was made with formulations such as amino-alcohols, e.g. delmopinol hydrochloride, which may selectively interfere with the microbial colonization of exposed tooth surfaces
0003-9969/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 3 - 9 9 6 9 ( 9 8 ) 0 0 0 3 2 - 6
560
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
(Simonsson et al., 1991) without markedly in¯uencing the oral ecological system (Rundegren et al., 1992). Basically, an antiadhesive eect can be ascribed to an agent that coats a tooth surface and thereby prevents bacteria attaching to that surface. A widely applied method of evaluating the eect of mouthrinses on the oral micro¯ora is to determine the colony-forming units in saliva (Addy et al., 1991; Jenkins et al., 1994). Even if extensive culture methods are employed, plate counts only measure colony-forming units not individual viable cells but give insight into the viable portion of the oral micro¯ora. However, colony-forming units do not provide reliable data on the number of vital and possibly non-viable bacteria and of those killed by chemotherapeutics. This might be overcome by marking vital and dead micro-organisms with ¯uorescent stains to evaluate the percentage of living and dead micro-organisms (Ziegler et al., 1975; Netuschil et al., 1989; Weiger et al., 1995; 1997). In addition to recording the microbial vitality in saliva, the determination of the total number of adherent bacteria by dark®eld microscopy together with their vitality status during the early stages of plaque development might help to discriminate between potential antibacterial and antiadhesive eects of mouthrinses. In this study an approach is presented that may allow the dierentiation between antibacterial and antiadhesive properties of mouthrinses in vivo. An amine ¯uoride/stannous ¯uoride preparation was used as an exemplary test substance to elucidate its mode of action. 2. Material and methods 2.1. Participants Twelve healthy adults (mean age 25 years; range 21± 28 years) were recruited for this study. They displayed no or only minor signs of gingivitis and no carious lesions could be diagnosed clinically. The number of ®lled tooth surfaces ranged from 0 to 33 (median 8.5). The participants had taken no antibiotics or medicaments known to aect the oral micro¯ora for at least 3 months before the experiment, and took none during it. 2.2. Study design In the preparatory phase, the 12 volunteers were subjected to a professional tooth cleaning. They were requested not to take any mouthwashes or related agents having antimicrobial eects on the oral micro¯ora during the entire experimental phase. In a crossover design they refrained from all oral-hygiene measures for four separate periods of 2 4 h and
2 72 h. Between these periods they maintained optimal plaque control with a toothbrush, dental ¯oss and ¯uoride-free dentifrice (Aronal forte1, Wybert, LoÈrrach, Germany). The wash-out time between the intervals was at least 7 days. The participants were asked to present themselves in the morning without having eaten. Under supervision they meticulously brushed their teeth with a toothbrush and dental ¯oss without any dentifrice. Finally, they rinsed with about 10 ml tap water. One and a half hours before the beginning of each experimental interval (ÿ1.5 h) stimulated whole saliva (17.5 ml) was collected by paran-wax stimulation. All intact vestibular surfaces of the teeth 16 to 26 selected for plaque sampling were cleaned with a smooth rubber cup without pumice and planed with a sterile curette. Every second surface was covered with an inert plastic ®lm (Plast Spray1). This ®lm sets within a few seconds, leaving a transparent covering on the surfaces (RoÈnstroÈm et al., 1975; Brecx et al., 1981, 1994). The participants then rinsed with either 10 ml of 0.9% NaCl ( placebo) or 10 ml of a test solution (Meridol1: 125 parts/106 from amine ¯uoride and 125 parts/106 from stannous ¯uoride; Wybert, LoÈrrach, Germany) for 1 min under supervision. At baseline (0 h) the uncovered tooth surfaces were carefully cleaned with a smooth rubber cup and then rinsed with 0.9% NaCl to remove the bacteria that had attached to these areas during the preceding 1.5 h. Finally, the plastic ®lms were removed from the other surfaces and the saliva samples taken. In the following text the surfaces temporarily covered with plastic ®lms are designated as protected sites in contrast to the unprotected sites, which were directly exposed to 0.9% NaCl or the test solution during rinsing. At the end of each experimental interval (4 or 72 h), paran-wax stimulated whole saliva was sampled (17.5 ml) and supragingival plaque separately collected as completely as possible with a sterile curette from the protected and the unprotected surfaces. In addition, stimulated saliva was collected and supragingival plaque gathered from 6 out of the 12 selected tooth surfaces after 24 h within the 72-h period. The 72-h-old plaque was gathered from the remaining test surfaces. A removable splint was used to mark the vestibular tooth areas from which plaque was taken. This technique allowed the sampling of plaque from de®ned tooth sites in a standardized and reproducible manner, and the determination of the size of these areas in mm2 (Weiger et al., 1992, 1995). The study design is schematically illustrated in Fig. 1. The participants were not informed about the kind of mouthrinse they were using. At the time of the microbiological analysis of the plaque and saliva samples the laboratory technician was not aware of the kind of the rinsing solution applied.
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
561
slide and analysed under a photomicroscope (Zeiss, Oberkochen, Germany) with an ultraviolet excitation ®lter (wavelength 450±490 nm). At a magni®cation of 312 , the percentage of living micro¯ora was evaluated in 125 small squares of a counting grid. For each sample the microbial vitality in each of the 125 squares was rated as 0%, 210%, 230%, 250%, 270%, 290% or 100%. The mean of the recorded values represented the microbial vitality of a speci®c saliva sample (Weiger et al., 1997). 2.4. Microbiological analysis of the plaque
Fig. 1. Study design. The participants rinsed with either the test solution Meridol1 or 0.9% NaCl as placebo.
2.3. Microbiological analysis of the saliva All saliva samples were processed immediately. They were dispersed in a Vortex mixer and sonicated for 10 s (TM cell disruptor W-225R; Heat Systems± Ultrasonics, Inc., Plainview, Texas, U.S.A.). 2.3.1. Total bacterial counts per ml saliva Of the original sample, 50 ml were serially diluted in physiological saline. A portion of 2.5 ml (10ÿ1 dilution) was transferred to a standard counting chamber to count the bacterial cells under the dark®eld microscope (Leitz Ortholux II, Wetzlar, Germany) at a magni®cation of 500 . BCs represents the sum of all living and dead micro-organisms present in the speci®c sample. 2.3.2. Percentage of vital bacteria in saliva The remaining saliva sample was ®ltered with ®rstly a 70-mm ®lter (Falcon, Cellstrain-Filter No. 2350, Becton-Dickinson) and then an 8-mm ®lter (SpectraMesh; Novodirect GmbH, Kehl, Germany) to separate the bacterial population from the epithelial cells and remnants. A portion (1 ml) of the ®ltrate was centrifuged for 7 min and the pellet resuspended in Schaedler broth (Becton-Dickinson, No. 12191) to eliminate the background ¯uorescence frequently observed pre-experimentally. This washing was done twice. The salivary bacterial cells were subsequently harvested by centrifugation and suspended in 500 ml staining solution prepared from a stock solution containing ¯uorescein diacetate and ethidium bromide, which were used to dierentiate between vital and irreversibly damaged (dead) salivary micro-organisms (Netuschil et al., 1989). The stained saliva ®ltrate was again centrifuged and the supernatant carefully suctioned o. A portion (3.5 ml) was transferred to a glass
Plaque was collected from each test tooth in turn with the same curette and transferred to a vial. The samples from the protected and the unprotected sites were put into separate vials. For this purpose the curette tip was rinsed with 1 ml of Schaedler broth and subsequently thoroughly agitated in the solution. The pooled samples were transported to the laboratory within 5 min and immediately processed. The methods to determine the total number of bacterial counts per mm2 tooth surface and the percentage of vital plaque bacteria were similar to those applied for saliva. The exact procedures have already been described in detail elsewhere (Weiger et al., 1992, 1995, 1997). 2.5. Statistical analysis The Wilcoxon test for matched pairs as a non-parametric approach was used to perform selected comparisons, with the individual being the statistical unit. A one-sided test was carried out in cases where the test solution appeared to lead directly or indirectly to a reduction in the percentage of viable bacteria in saliva, plaque bacterial count or percentage of plaque viable bacteria. As multiple comparisons were made (n = 14) the computed p-values were adjusted (=padj) according to Bonferroni±Holm (Holm, 1979). The level of statistical signi®cance a was set at 0.05. The analysis was restricted to the comparisons that are essential to revealing an antibacterial and/or antiadhesive eect of the test solution. Thus the percentage of vital bacteria in saliva was compared between samples 1.5 h before the rinse and at time 0, and between samples after placebo rinses and after test solution at 0, 4 and 24 h. The total bacterial counts in plaque were compared between placebo and test solutions on both protected and unprotected teeth, as were the percentages of viable bacteria in each plaque sample. In order to compare plaque formed under various conditions in the same individual the viable bacteria percentages from protected surfaces were subtracted from those for unprotected surfaces and the plaque bacterial counts/mm2 were expressed as a ratio of unprotected to protected surfaces.
562
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
3. Results All saliva (n = 156) and plaque samples (n = 144) could be completely analysed. In Figs 2±7 the results are illustrated by box and whisker plots showing the cumulative frequency of the variable of interest. The
Fig. 5. Total bacterial counts in supragingival plaque (BCp). Plaque established after rinsing with 0.9% NaCl or Meridol1 was sampled from the temporarily protected (`f ') or the unprotected (`u') tooth areas at various times (4, 24, 72 h).
Fig. 2. Percentage of vital bacteria in stimulated whole saliva (PVBs) collected before (ÿ1.5 h) and after rinsing with Meridol1 or 0.9% NaCl (0, 4, 24, 72 h).
Fig. 6. Comparison of the percentages of the vital microorganisms in supragingival plaque (PVBp) formed under various conditions. The corresponding PVBp values from the same individual were subtracted [e.g. PVBp(`u')PVBp(`f ') = ÿ 5%] (`u': unprotected sites; `f ': protected sites; Mer: Meridol1; Pla: placebo). Fig. 3. Total bacterial counts in stimulated whole saliva (BCs) collected before (ÿ1.5 h) and after rinsing with Meridol1 or 0.9% NaCl (0, 4, 24, 72 h).
Fig. 4. Percentage of vital bacteria in supragingival plaque (PVBp). Plaque established after rinsing with 0.9% NaCl or Meridol1 was sampled from the temporarily protected (`f ') or the unprotected (`u') tooth areas at various times (4, 24, 72 h).
Fig. 7. Comparison of the total bacterial counts in supragingival plaque (BCp) formed under various conditions. The corresponding BCp values from the same individual were brought in relation [e.g. BCp(`u')/BCp (`f ') = 1.2] (`u': unprotected sites; `f ': protected sites; Mer: Meridol1; Pla: placebo).
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
box indicates the lower (25th percentile) and upper (75th percentile) quartils. The central line marks the median (50th percentile), the whiskers represent the 10th and 90th percentiles. 3.1. Saliva The percentages of vital salivary micro-organisms exceeded 80% in nearly all samples taken before rinsing with one of the solutions (Fig. 2). Values dropped signi®cantly ( padjR0.001) to about 50±60% following rinsing with the test solution (0 h). These baseline values were signi®cantly ( padjR0.001) lower than those recorded after the placebo rinse. In the presence of 4h-old ( padj>0.05) and 24-h-old ( padj=0.017) plaque, the salivary viable bacteria percentages in the test-solution group often remained lower than the corresponding placebo values. After 72 h of plaque accumulation, the values in both groups reached those measured immediately before rinsing (ÿ1.5 h). Over time, neither the test solution nor the placebo markedly aected the total number of salivary bacteria (Fig. 3). Salivary bacterial counts were most often between 109 and 3 109. 3.2. Plaque After the placebo rinse no pronounced dierences could be detected when comparing the individual viable bacteria percentages in plaque of varying age formed on the protected sites with those on the unprotected areas (Figs 4 and 6a). Similar total numbers of micro-organisms were observed on these tooth surfaces at all times (Figs 5 and 7a). After the rinse with the test solution the percentage of vital bacteria in 4-h plaque grown on the unprotected surfaces was not signi®cantly ( padj>0.05) lower than that from the protected surfaces (Figs 4 and 6b). The plaque bacterial counts were similar in 4-h-old plaque on the unprotected and protected sites ( padj>0.05) (Figs 5 and 7b). On the protected sites the 4-h-old plaque harboured a signi®cantly ( padj=0.018) lower percentage of vital bacteria following application of the test solution (Figs 4 and 6c). Reduced levels could also be detected in more than half of the 24-h-old plaque samples. Although the tooth surfaces were temporarily protected, the test solution exerted only a limited in¯uence on plaque bacterial count in 4-h plaque in comparison to the placebo. Bacterial counts were diminished in some of these samples ( padj>0.05) (Figs 5 and 7c). On the unprotected surfaces the percentage of vital micro-organisms in 4-h-old plaque varied between 15 and 30% almost without exception (Fig. 4). With increasing plaque age these values rose to 60±90%. After the use of the test solution, percentages of viable bacteria in 4-h and 24-h-old plaque were signi®cantly
563
lower than the corresponding values in the placebo group (4 h: padj=0.05; 24 h: padj=0.03) (Figs 4 and 6d). The test solution did not have a distinct eect on the proportion of viable bacteria in 72-h-old plaque in comparison with the placebo values. In 4-h-old plaque originating from the unprotected areas, bacterial counts were frequently lower after the test solution than after the placebo ( padj>0.05) (Figs 5 and 7d). After 72 h of plaque accumulation, however, plaque bacteria counts exceeded 107 irrespective of the formulation used. 4. Discussion This investigation focused on plaque regrowth on clinically clean tooth surfaces following a single rinse with an amine ¯uoride/stannous ¯uoride solution. The amine ion acts not only as antibacterial compound but also as detergent (Schmid, 1983). Thus the amine ions may interfere with the early microbial colonization on tooth surfaces as a result of an antiadhesive eect. An approach to revealing the potential mode and duration of action of this mouthrinse in vivo was developed, particularly with regard to potential antibacterial and antiadhesive eects. For this purpose, newly established plaque of varying age formed on either unprotected or temporarily protected tooth areas was compared, and the microbial vitality in whole saliva was simultaneously monitored. Vestibular tooth surfaces were selected for plaque sampling as these are easily accessible and come into close contact with the agent. Supragingival plaque was collected from sites adjacent to primarily healthy gingival margins as the presence of gingival in¯ammation promotes early plaque growth (Brecx et al., 1983; Ramberg et al., 1996; Daly and High®eld, 1996). In order to avoid carry-over eects from the preceding treatments, particularly from the test solution, the wash-out times between the various experimental periods were at least 7 days (Newcombe et al., 1995). Half of the selected tooth areas were protected by plastic ®lms to avoid direct contact with the agent during rinsing. This enabled us to examine (1) whether direct exposure to the mouthrinse was a factor contributing to plaque inhibition, (2) whether an oral reservoir of the test solution was present, and (3) to what extent the desorption of components of the test solution from oral surfaces in¯uenced supragingival plaque regrowth. Certainly, instead of the plastic ®lms, the use of removable enamel inserts integrated in appliances would be an alternative (Jenkins et al., 1988). The growth pattern of the plaque formed on inserts, however, may not be comparable with that on natural tooth surfaces as they have usually to be placed at sites with dierent local environmental conditions (e.g. in the palate).
564
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
The plastic ®lm attached relatively ®rmly to the tooth surface as a macroscopically even layer. It is unlikely that during the short exposure to the rinse, ingredients of the test solution could diuse through the ®lm. Another important question concerns the possible consequences of the plastic-®lm application on enamel surfaces and on plaque formation. Obviously, both the protected and the unprotected tooth areas accumulated similar numbers of bacteria at all times after the placebo rinse. Likewise the proportions of vital micro-organisms present on these tooth surfaces were not very dierent. The results of RoÈnstroÈm et al. (1975, 1977) and Netuschil et al. (1991) indirectly provide evidence of the inert character of the plastic ®lm itself. They demonstrated that the pattern of early plaque formation and the vitality of the plaque ¯ora grown on such plastic ®lms are similar to those on enamel. The ®rst events of microbial colonization after exposure of a clean tooth surface to saliva are governed primarily by adhesion of salivary bacteria (for review, see Quirynen and Bollen, 1995). The proliferation of viable micro-organisms on the tooth surface leads to elevated bacterial counts only at later stages of plaque growth (e.g. after 24 or 72 h) (Brecx et al., 1983). As both types of surfacesÐunprotected or temporarily protectedÐprovide the same conditions for initial microbial colonization, the comparison of the total numbers of vital and dead micro-organisms initially adherent (e.g. after 4 h) on these areas following rinsing with the test solution may demonstrate whether or not the mouthrinse exerts an antiadhesive eect. Such an eect should reduce the plaque bacterial count on the unprotected sites at 4 h while the percentage of adherent vital cells remains close to constant. Our results suggest that direct exposure to the test solution yields similar bacterial counts during the initial stages of plaque formation provided that the number of potential microbial colonizers from saliva is similar at the various sites selected for plaque sampling. Thus the test solution did not demonstrate an antiadhesive eect. The percentage of plaque viable bacteria also did not dier at 4 h, indicating that the adsorption of some ingredients of the test solution onto or into the outermost layers of the unprotected tooth surfaces during rinsing (Tinano et al., 1976; Perdok et al., 1988; Busscher et al., 1988) did not contribute to the killing of already adherent bacteria. Yet, it cannot be denied that some active substances adsorbed onto these tooth areas might be polished o by the careful cleaning of the unprotected surfaces at baseline (0 h). However, this procedure was necessary to remove the bacteria attached to these tooth surfaces during the elapsed 1.5 h. By comparing the plaque formed on the protected tooth areas under the in¯uence of either the test sol-
ution or physiological saline, we found that the 4-h-old plaque of the test group harboured a lower percentage of vital micro-organisms. This dierence is possibly related to a change in the microbial vitality in saliva. As seen in Fig. 3, most of the salivary bacteria (80±95%) were vital before the rinse (ÿ1.5 h). After rinsing with the test solution the percentage of salivary bacteria dropped signi®cantly to about 50±65% within the ®rst 1.5 h. In many samples these values remained lower than those recorded for the placebo during the following 24 h. If the bacterial composition of whole saliva is a rough mirror of that in the mouth, the recorded variables for saliva may further elucidate the underlying mechanisms responsible for the reduced percentage of plaque viable bacteria in 4-h-old plaque following the use of the test solution. First, lower levels of vital salivary bacteria may be available for initial adhesion to the tooth surfaces, owing to the described antibacterial eect of the test solution on the salivary micro¯ora. Second, desorbed ingredients of the formulation may have a direct in¯uence on the vitality of the microorganisms already present. Third, some ingredients may adsorb to the originally protected tooth surfaces even after the removal of the plastic ®lms and interfere with the initial microbial colonization. Finally, the overall eect on de novo plaque formation is substantiated by the comparison of dental plaque of varying age formed on the unprotected sites after either Meridol1 or placebo rinses. There is evidence that the test solution reduces the microbial vitality in early supragingival plaque. Likewise the percentage of salivary micro-organisms in saliva was lowered for a few hours after the use of the formulation, indicating that the ecological balance of the salivary micro¯ora was disturbed for this period of time. The model presented here allows us to decode the mode of action of a mouthrinse. The amine ¯uoride/ stannous ¯uoride solution, for example, exerts a short antibacterial eect but no clear antiadhesive eect. As the whole product was tested in this study, it remains speculative to ascribe this antibacterial eect to one or both of its active ingredients, i.e., amine ¯uoride and stannous ¯uoride.
Acknowledgements This study was supported by GABA International, Basel, Switzerland. We thank Mrs Carmen BuckleyGoÈrbing for her assistance in preparing the manuscript.
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
References Adams, D., Addy, M., 1994. Mouthrinses. Adv. Dent. Res. 8, 291±301. Addy, M., Jenkins, S., Newcombe, R., 1991. The eect of some chlorhexidine-containing mouthrinses on salivary bacterial counts. J. Clin. Periodontol. 18, 90±93. Brecx, M., RoÈnstroÈm, A., Theilade, J., AttstroÈm, R., 1981. Early formation of dental plaque on plastic ®lms. 2. Electron microscopic observations. J. Periodont. Res. 16, 213±227. Brecx, M., Theilade, J., AttstroÈm, R., 1983. An ultrastructural quantitative study of the signi®cance of microbial multiplication during early dental plaque growth. J. Periodont. Res. 18, 177±186. Brecx, M., Winkler, M., Netuschil, L., 1994. Human dental plaque formation on plastic ®lms. A quantitative SEM study. J. West. Soc. Perio. 42, 77±80. Brecx, M., 1997. Strategies and agents in supragingival plaque control: a review. Periodontol. 15, 100±108. Busscher, H.J., Uyen, H.M., de Jong, H.P., Arends, J., 1988. Adsorption of amine ¯uorides on human enamel. J. Dent. 16, 166±171. Daly, C.G., High®eld, J.E., 1996. Eect of localized experimental gingivitis on early supragingival plaque accumulation. J. Clin. Periodontol. 23, 160±164. Holm, S., 1979. A simple sequentially rejective multiple test procedure. Scand. J. Statist. 6, 65±70. Jenkins, S., Addy, M., Wade, W., 1988. The mechanism of action of chlorhexidine. A study of plaque growth on enamel inserts in vivo. J. Clin. Periodontol. 15, 415±424. Jenkins, S., Addy, M., Wade, W., Newcombe, R., 1994. The magnitude and duration of the eects of some mouthrinse products on salivary bacterial counts. J. Clin. Periodontol. 21, 397±401. Netuschil, L., Reich, E., Brecx, M., 1989. Direct measurement of the bactericidal eect of chlorhexidine on human dental plaque. J. Clin. Periodontol. 16, 484±488. Netuschil, L., Barthold, G., Brecx, M., 1991. Vital¯uoreszenzfaÈmit Fluoresceindiacetat und Ethidiumbromid. Z. Stomatol. 88, 147±153. Newcombe, R.G., Addy, M., McKeown, S., 1995. Residual eect of chlorhexidine gluconate in 4-day plaque regrowth crossover trials, and its implication for study design. J. Periodont. Res. 30, 319±324. Perdok, J.F., Busscher, H.J., Weerkamp, A.H., Arends, J., 1988. The eect of an amine¯uoride-stannous ¯uoride-con-
565
taining mouthrinse on enamel surface free energy and the development of plaque and gingivitis. Clin. Prevent. Dent. 10, 3±9. Quirynen, M., Bollen, C.M.L., 1995. The in¯uence of surface roughness and surface-free energy on supra- and subgingival plaque formation. J. Clin. Periodontol. 22, 1±4. Ramberg, P., Furuichi, Y., Volpe, A.R., Gaar, A., Lindhe, J., 1996. The eects of antimicrobial mouthrinses on de novo plaque formation at sites with healthy and in¯amed gingiva. J. Clin. Periodontol. 23, 7±11. RoÈnstroÈm, A., AttstroÈm, R., Egelberg, J., 1975. Early formation of dental plaque on plastic ®lms. J. Periodont. Res. 10, 28±35. RoÈnstroÈm, A., Edwardsson, S., AttstroÈm, R., 1977. Streptococcus sanguis and Streptococcus salivarius in early plaque formation on plastic ®lms. J. Periodont. Res. 12, 331±339. Rundegren, J., Hvid, E.B., Johansson, M., AstroÈm, M., 1992. Eect of 4 days of mouth rinsing with delmopinol or chlorhexidine on the vitality of plaque bacteria. J. Clin. Periodontol. 19, 322±325. Schmid, H., 1983. Chemie und Ober¯aÈchenwirkungen der Amin¯uoride. Dtsch. ZahnaÈrztl. Z. 38, S1±S9. Simonsson, T., Arnebrant, T., Peterson, L., 1991. The eect of delmopinol on salivary pellicle, the wettability of tooth surfaces in vivo and bacterial surfaces in vitro. Biofouling 3, 251±260. Tinano, N., Brady, J.M., Gross, A., 1976. The eect of NaF and SnF2 mouthrinses on bacterial colonization of tooth enamel: TEM and SEM studies. Caries Res. 10, 415±426. Weiger, R., Netuschil, L., Brecx, M., 1992. Relationship between bacterial counts, microbial vitality and the accumulation of supragingival plaque in humans. J. Periodont. Res. 27, 575±580. Weiger, R., Netuschil, L., von Ohle, C., Brecx, M., 1995. Microbial vitality of supragingival dental plaque during initial stages of experimental gingivitis in humans. J. Periodont. Res. 30, 204±209. Weiger, R., von Ohle, C., Decker, E., Axmann-Krcmar, D., Netuschil, L., 1997. Vital micro-organisms in early supragingival dental plaque and in stimulated human saliva. J. Periodont. Res. 32, 233±240. Ziegler, B., Ziegler, E., Witzenhausen, R., 1975. Nachweis der StowechselaktivitaÈt von Mikroorganismen durch VitalFluorchromierung mit 3',6'-Diacetyl¯uorescein. Zbl. Bakt. Hyg. 230, 252±264.
PERGAMON
Archives of Oral Biology 43 (1998) 559±565
ORAL BIOLOGY
An approach to dierentiate between antibacterial and antiadhesive eects of mouthrinses in vivo R. Weiger *, L. Netuschil, T. Wester-Ebbinghaus, M. Brecx Department of Conservative Dentistry, School of Dental Medicine, University of TuÈbingen, Osianderstr. 2±8, 72076 TuÈbingen, Germany Accepted 3 March 1998
Abstract An experimental set-up allowing dierentiation in vivo between antibacterial and antiadhesive properties of mouthrinses is described. The percentage of vital bacteria (=microbial vitality) and the bacterial counts were microscopically evaluated in saliva and in supragingival dental plaque both collected simultaneously at various times during de novo plaque formation. In a cross-over design, 12 healthy participants refrained from all oral hygiene for four separate periods of 2 4 h and 2 72 h after having rinsed with either an amine ¯uoride/stannous ¯uoride solution (Meridol1) or 0.9% NaCl (placebo). Stimulated whole saliva was collected before and after the rinse. Together with whole-saliva samples, representative 4, 24 and 72-h-old plaque samples were separately taken from de®ned vestibular tooth surfaces that had been either exposed to the mouthrinse (unprotected sites) or temporarily covered with inert plastic ®lms (protected sites) during rinsing. The pooled plaque and saliva were stained with ¯uorescent dyes to dierentiate vital from dead micro-organisms which permitted the estimation of the percentages of vital bacteria. The total bacterial counts were quanti®ed under the dark®eld microscope. The Wilcoxon test was used for selected pairwise comparisons (a = 0.05). The percentage of vital bacteria in saliva fell signi®cantly from 80±95% to about 50±60% as a result of the antibacterial activity of the test solution. These baseline values and those found in the presence of 4 and 24-h-old plaque were frequently lower than those recorded after the placebo rinse. In comparison to the placebo, microbial vitality was signi®cantly reduced in early supragingival plaque formed on unprotected sites after applying the test solution. The similar total bacterial counts in 4-h-old plaque recorded after the use of the test solution on the unprotected and the protected areas did not point to an antiadhesive eect of the agent. It is concluded that this new experimental set-up allows decoding of the mode of action of a mouthrinse. # 1998 Elsevier Science Ltd. All rights reserved. Keywords: Mouthrinse; Dental plaque; Saliva; Amine ¯uoride; Stannous ¯uoride
1. Introduction Mouthrinses used daily should inhibit or suppress supragingival dental plaque formation and thereby prevent caries and gingivitis. A variety of dierent oral chemotherapeutics are now available, with chlorhexidine being the most ecient antiplaque agent (Adams
* Corresponding (49)7071-295656
author.
Tel.:
(49)7071-2983498;
fax:
and Addy, 1994; Brecx, 1997). Most of them have antibacterial properties, which explain their eect on plaque development. From a biological point of view, however, it is desirable to maintain the natural balance of the oral micro¯ora rather than to disturb it by the use of any antibacterial mouthrinse. For this reason, substances with no or only slight antimicrobial activity are sought. A ®rst approach was made with formulations such as amino-alcohols, e.g. delmopinol hydrochloride, which may selectively interfere with the microbial colonization of exposed tooth surfaces
0003-9969/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 3 - 9 9 6 9 ( 9 8 ) 0 0 0 3 2 - 6
560
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
(Simonsson et al., 1991) without markedly in¯uencing the oral ecological system (Rundegren et al., 1992). Basically, an antiadhesive eect can be ascribed to an agent that coats a tooth surface and thereby prevents bacteria attaching to that surface. A widely applied method of evaluating the eect of mouthrinses on the oral micro¯ora is to determine the colony-forming units in saliva (Addy et al., 1991; Jenkins et al., 1994). Even if extensive culture methods are employed, plate counts only measure colony-forming units not individual viable cells but give insight into the viable portion of the oral micro¯ora. However, colony-forming units do not provide reliable data on the number of vital and possibly non-viable bacteria and of those killed by chemotherapeutics. This might be overcome by marking vital and dead micro-organisms with ¯uorescent stains to evaluate the percentage of living and dead micro-organisms (Ziegler et al., 1975; Netuschil et al., 1989; Weiger et al., 1995; 1997). In addition to recording the microbial vitality in saliva, the determination of the total number of adherent bacteria by dark®eld microscopy together with their vitality status during the early stages of plaque development might help to discriminate between potential antibacterial and antiadhesive eects of mouthrinses. In this study an approach is presented that may allow the dierentiation between antibacterial and antiadhesive properties of mouthrinses in vivo. An amine ¯uoride/stannous ¯uoride preparation was used as an exemplary test substance to elucidate its mode of action. 2. Material and methods 2.1. Participants Twelve healthy adults (mean age 25 years; range 21± 28 years) were recruited for this study. They displayed no or only minor signs of gingivitis and no carious lesions could be diagnosed clinically. The number of ®lled tooth surfaces ranged from 0 to 33 (median 8.5). The participants had taken no antibiotics or medicaments known to aect the oral micro¯ora for at least 3 months before the experiment, and took none during it. 2.2. Study design In the preparatory phase, the 12 volunteers were subjected to a professional tooth cleaning. They were requested not to take any mouthwashes or related agents having antimicrobial eects on the oral micro¯ora during the entire experimental phase. In a crossover design they refrained from all oral-hygiene measures for four separate periods of 2 4 h and
2 72 h. Between these periods they maintained optimal plaque control with a toothbrush, dental ¯oss and ¯uoride-free dentifrice (Aronal forte1, Wybert, LoÈrrach, Germany). The wash-out time between the intervals was at least 7 days. The participants were asked to present themselves in the morning without having eaten. Under supervision they meticulously brushed their teeth with a toothbrush and dental ¯oss without any dentifrice. Finally, they rinsed with about 10 ml tap water. One and a half hours before the beginning of each experimental interval (ÿ1.5 h) stimulated whole saliva (17.5 ml) was collected by paran-wax stimulation. All intact vestibular surfaces of the teeth 16 to 26 selected for plaque sampling were cleaned with a smooth rubber cup without pumice and planed with a sterile curette. Every second surface was covered with an inert plastic ®lm (Plast Spray1). This ®lm sets within a few seconds, leaving a transparent covering on the surfaces (RoÈnstroÈm et al., 1975; Brecx et al., 1981, 1994). The participants then rinsed with either 10 ml of 0.9% NaCl ( placebo) or 10 ml of a test solution (Meridol1: 125 parts/106 from amine ¯uoride and 125 parts/106 from stannous ¯uoride; Wybert, LoÈrrach, Germany) for 1 min under supervision. At baseline (0 h) the uncovered tooth surfaces were carefully cleaned with a smooth rubber cup and then rinsed with 0.9% NaCl to remove the bacteria that had attached to these areas during the preceding 1.5 h. Finally, the plastic ®lms were removed from the other surfaces and the saliva samples taken. In the following text the surfaces temporarily covered with plastic ®lms are designated as protected sites in contrast to the unprotected sites, which were directly exposed to 0.9% NaCl or the test solution during rinsing. At the end of each experimental interval (4 or 72 h), paran-wax stimulated whole saliva was sampled (17.5 ml) and supragingival plaque separately collected as completely as possible with a sterile curette from the protected and the unprotected surfaces. In addition, stimulated saliva was collected and supragingival plaque gathered from 6 out of the 12 selected tooth surfaces after 24 h within the 72-h period. The 72-h-old plaque was gathered from the remaining test surfaces. A removable splint was used to mark the vestibular tooth areas from which plaque was taken. This technique allowed the sampling of plaque from de®ned tooth sites in a standardized and reproducible manner, and the determination of the size of these areas in mm2 (Weiger et al., 1992, 1995). The study design is schematically illustrated in Fig. 1. The participants were not informed about the kind of mouthrinse they were using. At the time of the microbiological analysis of the plaque and saliva samples the laboratory technician was not aware of the kind of the rinsing solution applied.
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
561
slide and analysed under a photomicroscope (Zeiss, Oberkochen, Germany) with an ultraviolet excitation ®lter (wavelength 450±490 nm). At a magni®cation of 312 , the percentage of living micro¯ora was evaluated in 125 small squares of a counting grid. For each sample the microbial vitality in each of the 125 squares was rated as 0%, 210%, 230%, 250%, 270%, 290% or 100%. The mean of the recorded values represented the microbial vitality of a speci®c saliva sample (Weiger et al., 1997). 2.4. Microbiological analysis of the plaque
Fig. 1. Study design. The participants rinsed with either the test solution Meridol1 or 0.9% NaCl as placebo.
2.3. Microbiological analysis of the saliva All saliva samples were processed immediately. They were dispersed in a Vortex mixer and sonicated for 10 s (TM cell disruptor W-225R; Heat Systems± Ultrasonics, Inc., Plainview, Texas, U.S.A.). 2.3.1. Total bacterial counts per ml saliva Of the original sample, 50 ml were serially diluted in physiological saline. A portion of 2.5 ml (10ÿ1 dilution) was transferred to a standard counting chamber to count the bacterial cells under the dark®eld microscope (Leitz Ortholux II, Wetzlar, Germany) at a magni®cation of 500 . BCs represents the sum of all living and dead micro-organisms present in the speci®c sample. 2.3.2. Percentage of vital bacteria in saliva The remaining saliva sample was ®ltered with ®rstly a 70-mm ®lter (Falcon, Cellstrain-Filter No. 2350, Becton-Dickinson) and then an 8-mm ®lter (SpectraMesh; Novodirect GmbH, Kehl, Germany) to separate the bacterial population from the epithelial cells and remnants. A portion (1 ml) of the ®ltrate was centrifuged for 7 min and the pellet resuspended in Schaedler broth (Becton-Dickinson, No. 12191) to eliminate the background ¯uorescence frequently observed pre-experimentally. This washing was done twice. The salivary bacterial cells were subsequently harvested by centrifugation and suspended in 500 ml staining solution prepared from a stock solution containing ¯uorescein diacetate and ethidium bromide, which were used to dierentiate between vital and irreversibly damaged (dead) salivary micro-organisms (Netuschil et al., 1989). The stained saliva ®ltrate was again centrifuged and the supernatant carefully suctioned o. A portion (3.5 ml) was transferred to a glass
Plaque was collected from each test tooth in turn with the same curette and transferred to a vial. The samples from the protected and the unprotected sites were put into separate vials. For this purpose the curette tip was rinsed with 1 ml of Schaedler broth and subsequently thoroughly agitated in the solution. The pooled samples were transported to the laboratory within 5 min and immediately processed. The methods to determine the total number of bacterial counts per mm2 tooth surface and the percentage of vital plaque bacteria were similar to those applied for saliva. The exact procedures have already been described in detail elsewhere (Weiger et al., 1992, 1995, 1997). 2.5. Statistical analysis The Wilcoxon test for matched pairs as a non-parametric approach was used to perform selected comparisons, with the individual being the statistical unit. A one-sided test was carried out in cases where the test solution appeared to lead directly or indirectly to a reduction in the percentage of viable bacteria in saliva, plaque bacterial count or percentage of plaque viable bacteria. As multiple comparisons were made (n = 14) the computed p-values were adjusted (=padj) according to Bonferroni±Holm (Holm, 1979). The level of statistical signi®cance a was set at 0.05. The analysis was restricted to the comparisons that are essential to revealing an antibacterial and/or antiadhesive eect of the test solution. Thus the percentage of vital bacteria in saliva was compared between samples 1.5 h before the rinse and at time 0, and between samples after placebo rinses and after test solution at 0, 4 and 24 h. The total bacterial counts in plaque were compared between placebo and test solutions on both protected and unprotected teeth, as were the percentages of viable bacteria in each plaque sample. In order to compare plaque formed under various conditions in the same individual the viable bacteria percentages from protected surfaces were subtracted from those for unprotected surfaces and the plaque bacterial counts/mm2 were expressed as a ratio of unprotected to protected surfaces.
562
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
3. Results All saliva (n = 156) and plaque samples (n = 144) could be completely analysed. In Figs 2±7 the results are illustrated by box and whisker plots showing the cumulative frequency of the variable of interest. The
Fig. 5. Total bacterial counts in supragingival plaque (BCp). Plaque established after rinsing with 0.9% NaCl or Meridol1 was sampled from the temporarily protected (`f ') or the unprotected (`u') tooth areas at various times (4, 24, 72 h).
Fig. 2. Percentage of vital bacteria in stimulated whole saliva (PVBs) collected before (ÿ1.5 h) and after rinsing with Meridol1 or 0.9% NaCl (0, 4, 24, 72 h).
Fig. 6. Comparison of the percentages of the vital microorganisms in supragingival plaque (PVBp) formed under various conditions. The corresponding PVBp values from the same individual were subtracted [e.g. PVBp(`u')PVBp(`f ') = ÿ 5%] (`u': unprotected sites; `f ': protected sites; Mer: Meridol1; Pla: placebo). Fig. 3. Total bacterial counts in stimulated whole saliva (BCs) collected before (ÿ1.5 h) and after rinsing with Meridol1 or 0.9% NaCl (0, 4, 24, 72 h).
Fig. 4. Percentage of vital bacteria in supragingival plaque (PVBp). Plaque established after rinsing with 0.9% NaCl or Meridol1 was sampled from the temporarily protected (`f ') or the unprotected (`u') tooth areas at various times (4, 24, 72 h).
Fig. 7. Comparison of the total bacterial counts in supragingival plaque (BCp) formed under various conditions. The corresponding BCp values from the same individual were brought in relation [e.g. BCp(`u')/BCp (`f ') = 1.2] (`u': unprotected sites; `f ': protected sites; Mer: Meridol1; Pla: placebo).
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
box indicates the lower (25th percentile) and upper (75th percentile) quartils. The central line marks the median (50th percentile), the whiskers represent the 10th and 90th percentiles. 3.1. Saliva The percentages of vital salivary micro-organisms exceeded 80% in nearly all samples taken before rinsing with one of the solutions (Fig. 2). Values dropped signi®cantly ( padjR0.001) to about 50±60% following rinsing with the test solution (0 h). These baseline values were signi®cantly ( padjR0.001) lower than those recorded after the placebo rinse. In the presence of 4h-old ( padj>0.05) and 24-h-old ( padj=0.017) plaque, the salivary viable bacteria percentages in the test-solution group often remained lower than the corresponding placebo values. After 72 h of plaque accumulation, the values in both groups reached those measured immediately before rinsing (ÿ1.5 h). Over time, neither the test solution nor the placebo markedly aected the total number of salivary bacteria (Fig. 3). Salivary bacterial counts were most often between 109 and 3 109. 3.2. Plaque After the placebo rinse no pronounced dierences could be detected when comparing the individual viable bacteria percentages in plaque of varying age formed on the protected sites with those on the unprotected areas (Figs 4 and 6a). Similar total numbers of micro-organisms were observed on these tooth surfaces at all times (Figs 5 and 7a). After the rinse with the test solution the percentage of vital bacteria in 4-h plaque grown on the unprotected surfaces was not signi®cantly ( padj>0.05) lower than that from the protected surfaces (Figs 4 and 6b). The plaque bacterial counts were similar in 4-h-old plaque on the unprotected and protected sites ( padj>0.05) (Figs 5 and 7b). On the protected sites the 4-h-old plaque harboured a signi®cantly ( padj=0.018) lower percentage of vital bacteria following application of the test solution (Figs 4 and 6c). Reduced levels could also be detected in more than half of the 24-h-old plaque samples. Although the tooth surfaces were temporarily protected, the test solution exerted only a limited in¯uence on plaque bacterial count in 4-h plaque in comparison to the placebo. Bacterial counts were diminished in some of these samples ( padj>0.05) (Figs 5 and 7c). On the unprotected surfaces the percentage of vital micro-organisms in 4-h-old plaque varied between 15 and 30% almost without exception (Fig. 4). With increasing plaque age these values rose to 60±90%. After the use of the test solution, percentages of viable bacteria in 4-h and 24-h-old plaque were signi®cantly
563
lower than the corresponding values in the placebo group (4 h: padj=0.05; 24 h: padj=0.03) (Figs 4 and 6d). The test solution did not have a distinct eect on the proportion of viable bacteria in 72-h-old plaque in comparison with the placebo values. In 4-h-old plaque originating from the unprotected areas, bacterial counts were frequently lower after the test solution than after the placebo ( padj>0.05) (Figs 5 and 7d). After 72 h of plaque accumulation, however, plaque bacteria counts exceeded 107 irrespective of the formulation used. 4. Discussion This investigation focused on plaque regrowth on clinically clean tooth surfaces following a single rinse with an amine ¯uoride/stannous ¯uoride solution. The amine ion acts not only as antibacterial compound but also as detergent (Schmid, 1983). Thus the amine ions may interfere with the early microbial colonization on tooth surfaces as a result of an antiadhesive eect. An approach to revealing the potential mode and duration of action of this mouthrinse in vivo was developed, particularly with regard to potential antibacterial and antiadhesive eects. For this purpose, newly established plaque of varying age formed on either unprotected or temporarily protected tooth areas was compared, and the microbial vitality in whole saliva was simultaneously monitored. Vestibular tooth surfaces were selected for plaque sampling as these are easily accessible and come into close contact with the agent. Supragingival plaque was collected from sites adjacent to primarily healthy gingival margins as the presence of gingival in¯ammation promotes early plaque growth (Brecx et al., 1983; Ramberg et al., 1996; Daly and High®eld, 1996). In order to avoid carry-over eects from the preceding treatments, particularly from the test solution, the wash-out times between the various experimental periods were at least 7 days (Newcombe et al., 1995). Half of the selected tooth areas were protected by plastic ®lms to avoid direct contact with the agent during rinsing. This enabled us to examine (1) whether direct exposure to the mouthrinse was a factor contributing to plaque inhibition, (2) whether an oral reservoir of the test solution was present, and (3) to what extent the desorption of components of the test solution from oral surfaces in¯uenced supragingival plaque regrowth. Certainly, instead of the plastic ®lms, the use of removable enamel inserts integrated in appliances would be an alternative (Jenkins et al., 1988). The growth pattern of the plaque formed on inserts, however, may not be comparable with that on natural tooth surfaces as they have usually to be placed at sites with dierent local environmental conditions (e.g. in the palate).
564
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
The plastic ®lm attached relatively ®rmly to the tooth surface as a macroscopically even layer. It is unlikely that during the short exposure to the rinse, ingredients of the test solution could diuse through the ®lm. Another important question concerns the possible consequences of the plastic-®lm application on enamel surfaces and on plaque formation. Obviously, both the protected and the unprotected tooth areas accumulated similar numbers of bacteria at all times after the placebo rinse. Likewise the proportions of vital micro-organisms present on these tooth surfaces were not very dierent. The results of RoÈnstroÈm et al. (1975, 1977) and Netuschil et al. (1991) indirectly provide evidence of the inert character of the plastic ®lm itself. They demonstrated that the pattern of early plaque formation and the vitality of the plaque ¯ora grown on such plastic ®lms are similar to those on enamel. The ®rst events of microbial colonization after exposure of a clean tooth surface to saliva are governed primarily by adhesion of salivary bacteria (for review, see Quirynen and Bollen, 1995). The proliferation of viable micro-organisms on the tooth surface leads to elevated bacterial counts only at later stages of plaque growth (e.g. after 24 or 72 h) (Brecx et al., 1983). As both types of surfacesÐunprotected or temporarily protectedÐprovide the same conditions for initial microbial colonization, the comparison of the total numbers of vital and dead micro-organisms initially adherent (e.g. after 4 h) on these areas following rinsing with the test solution may demonstrate whether or not the mouthrinse exerts an antiadhesive eect. Such an eect should reduce the plaque bacterial count on the unprotected sites at 4 h while the percentage of adherent vital cells remains close to constant. Our results suggest that direct exposure to the test solution yields similar bacterial counts during the initial stages of plaque formation provided that the number of potential microbial colonizers from saliva is similar at the various sites selected for plaque sampling. Thus the test solution did not demonstrate an antiadhesive eect. The percentage of plaque viable bacteria also did not dier at 4 h, indicating that the adsorption of some ingredients of the test solution onto or into the outermost layers of the unprotected tooth surfaces during rinsing (Tinano et al., 1976; Perdok et al., 1988; Busscher et al., 1988) did not contribute to the killing of already adherent bacteria. Yet, it cannot be denied that some active substances adsorbed onto these tooth areas might be polished o by the careful cleaning of the unprotected surfaces at baseline (0 h). However, this procedure was necessary to remove the bacteria attached to these tooth surfaces during the elapsed 1.5 h. By comparing the plaque formed on the protected tooth areas under the in¯uence of either the test sol-
ution or physiological saline, we found that the 4-h-old plaque of the test group harboured a lower percentage of vital micro-organisms. This dierence is possibly related to a change in the microbial vitality in saliva. As seen in Fig. 3, most of the salivary bacteria (80±95%) were vital before the rinse (ÿ1.5 h). After rinsing with the test solution the percentage of salivary bacteria dropped signi®cantly to about 50±65% within the ®rst 1.5 h. In many samples these values remained lower than those recorded for the placebo during the following 24 h. If the bacterial composition of whole saliva is a rough mirror of that in the mouth, the recorded variables for saliva may further elucidate the underlying mechanisms responsible for the reduced percentage of plaque viable bacteria in 4-h-old plaque following the use of the test solution. First, lower levels of vital salivary bacteria may be available for initial adhesion to the tooth surfaces, owing to the described antibacterial eect of the test solution on the salivary micro¯ora. Second, desorbed ingredients of the formulation may have a direct in¯uence on the vitality of the microorganisms already present. Third, some ingredients may adsorb to the originally protected tooth surfaces even after the removal of the plastic ®lms and interfere with the initial microbial colonization. Finally, the overall eect on de novo plaque formation is substantiated by the comparison of dental plaque of varying age formed on the unprotected sites after either Meridol1 or placebo rinses. There is evidence that the test solution reduces the microbial vitality in early supragingival plaque. Likewise the percentage of salivary micro-organisms in saliva was lowered for a few hours after the use of the formulation, indicating that the ecological balance of the salivary micro¯ora was disturbed for this period of time. The model presented here allows us to decode the mode of action of a mouthrinse. The amine ¯uoride/ stannous ¯uoride solution, for example, exerts a short antibacterial eect but no clear antiadhesive eect. As the whole product was tested in this study, it remains speculative to ascribe this antibacterial eect to one or both of its active ingredients, i.e., amine ¯uoride and stannous ¯uoride.
Acknowledgements This study was supported by GABA International, Basel, Switzerland. We thank Mrs Carmen BuckleyGoÈrbing for her assistance in preparing the manuscript.
R. Weiger et al. / Archives of Oral Biology 43 (1998) 559±565
References Adams, D., Addy, M., 1994. Mouthrinses. Adv. Dent. Res. 8, 291±301. Addy, M., Jenkins, S., Newcombe, R., 1991. The eect of some chlorhexidine-containing mouthrinses on salivary bacterial counts. J. Clin. Periodontol. 18, 90±93. Brecx, M., RoÈnstroÈm, A., Theilade, J., AttstroÈm, R., 1981. Early formation of dental plaque on plastic ®lms. 2. Electron microscopic observations. J. Periodont. Res. 16, 213±227. Brecx, M., Theilade, J., AttstroÈm, R., 1983. An ultrastructural quantitative study of the signi®cance of microbial multiplication during early dental plaque growth. J. Periodont. Res. 18, 177±186. Brecx, M., Winkler, M., Netuschil, L., 1994. Human dental plaque formation on plastic ®lms. A quantitative SEM study. J. West. Soc. Perio. 42, 77±80. Brecx, M., 1997. Strategies and agents in supragingival plaque control: a review. Periodontol. 15, 100±108. Busscher, H.J., Uyen, H.M., de Jong, H.P., Arends, J., 1988. Adsorption of amine ¯uorides on human enamel. J. Dent. 16, 166±171. Daly, C.G., High®eld, J.E., 1996. Eect of localized experimental gingivitis on early supragingival plaque accumulation. J. Clin. Periodontol. 23, 160±164. Holm, S., 1979. A simple sequentially rejective multiple test procedure. Scand. J. Statist. 6, 65±70. Jenkins, S., Addy, M., Wade, W., 1988. The mechanism of action of chlorhexidine. A study of plaque growth on enamel inserts in vivo. J. Clin. Periodontol. 15, 415±424. Jenkins, S., Addy, M., Wade, W., Newcombe, R., 1994. The magnitude and duration of the eects of some mouthrinse products on salivary bacterial counts. J. Clin. Periodontol. 21, 397±401. Netuschil, L., Reich, E., Brecx, M., 1989. Direct measurement of the bactericidal eect of chlorhexidine on human dental plaque. J. Clin. Periodontol. 16, 484±488. Netuschil, L., Barthold, G., Brecx, M., 1991. Vital¯uoreszenzfaÈmit Fluoresceindiacetat und Ethidiumbromid. Z. Stomatol. 88, 147±153. Newcombe, R.G., Addy, M., McKeown, S., 1995. Residual eect of chlorhexidine gluconate in 4-day plaque regrowth crossover trials, and its implication for study design. J. Periodont. Res. 30, 319±324. Perdok, J.F., Busscher, H.J., Weerkamp, A.H., Arends, J., 1988. The eect of an amine¯uoride-stannous ¯uoride-con-
565
taining mouthrinse on enamel surface free energy and the development of plaque and gingivitis. Clin. Prevent. Dent. 10, 3±9. Quirynen, M., Bollen, C.M.L., 1995. The in¯uence of surface roughness and surface-free energy on supra- and subgingival plaque formation. J. Clin. Periodontol. 22, 1±4. Ramberg, P., Furuichi, Y., Volpe, A.R., Gaar, A., Lindhe, J., 1996. The eects of antimicrobial mouthrinses on de novo plaque formation at sites with healthy and in¯amed gingiva. J. Clin. Periodontol. 23, 7±11. RoÈnstroÈm, A., AttstroÈm, R., Egelberg, J., 1975. Early formation of dental plaque on plastic ®lms. J. Periodont. Res. 10, 28±35. RoÈnstroÈm, A., Edwardsson, S., AttstroÈm, R., 1977. Streptococcus sanguis and Streptococcus salivarius in early plaque formation on plastic ®lms. J. Periodont. Res. 12, 331±339. Rundegren, J., Hvid, E.B., Johansson, M., AstroÈm, M., 1992. Eect of 4 days of mouth rinsing with delmopinol or chlorhexidine on the vitality of plaque bacteria. J. Clin. Periodontol. 19, 322±325. Schmid, H., 1983. Chemie und Ober¯aÈchenwirkungen der Amin¯uoride. Dtsch. ZahnaÈrztl. Z. 38, S1±S9. Simonsson, T., Arnebrant, T., Peterson, L., 1991. The eect of delmopinol on salivary pellicle, the wettability of tooth surfaces in vivo and bacterial surfaces in vitro. Biofouling 3, 251±260. Tinano, N., Brady, J.M., Gross, A., 1976. The eect of NaF and SnF2 mouthrinses on bacterial colonization of tooth enamel: TEM and SEM studies. Caries Res. 10, 415±426. Weiger, R., Netuschil, L., Brecx, M., 1992. Relationship between bacterial counts, microbial vitality and the accumulation of supragingival plaque in humans. J. Periodont. Res. 27, 575±580. Weiger, R., Netuschil, L., von Ohle, C., Brecx, M., 1995. Microbial vitality of supragingival dental plaque during initial stages of experimental gingivitis in humans. J. Periodont. Res. 30, 204±209. Weiger, R., von Ohle, C., Decker, E., Axmann-Krcmar, D., Netuschil, L., 1997. Vital micro-organisms in early supragingival dental plaque and in stimulated human saliva. J. Periodont. Res. 32, 233±240. Ziegler, B., Ziegler, E., Witzenhausen, R., 1975. Nachweis der StowechselaktivitaÈt von Mikroorganismen durch VitalFluorchromierung mit 3',6'-Diacetyl¯uorescein. Zbl. Bakt. Hyg. 230, 252±264.