The effects of flossing with a chlorhexidine solution on interproximal gingivitis: a randomized controlled trial.

ABSTRACT

Background: Gingivitis is an inflammatory response of the gingival tissues to bacterial plaque that can be treated by brushing and flossing or rinsing with chlorhexidine. This study examined whether floss presoaked in chlorhexidine improved oral health relative to flossing alone. Methods: A 3-month, parallel, randomized control trial was conducted on 27 adults with a minimum of 10 bleeding sites, who were randomly assigned to a floss soaked in 0.12 per cent chlorhexidine or floss soaked in a placebo, quinine sulfate. Debridement and flossing instructions were performed at Week--1. Probing depth, bleeding on probing, gingival plaque, and stain indices were assessed at Weeks 0, 6, and 12. Flossing compliance was monitored by self-reports and length of floss used. Results: Flossing compliance was high for both groups. All subjects had statistically significant reductions in gingival indices scores (p < 0.0001). The chlorhexidine group had statistically significant reductions for probing depth at Week 6 (p = 0.03); the effect was more pronounced in shallow sites (probing depth < 4 mm) Week 6 (p = 0.01) and Week 12 (p = 0.01). The chlorhexidine group also had statistically significant reductions for bleeding on probing in subjects with moderate gingivitis (p = 0.01) and in all areas of the mouth (p = 0.01 anterior; p = 0.04 posterior). The two groups did not differ significantly for stain and plaque indices. Conclusion: Flossing with chlorhexidine reduces probing depths and bleeding on probing in subjects with moderate gingivitis compared to flossing alone.

RESUME

Contexte : La gingivite est une reaction inflammatoire des tissus gingivaux a la plaque bacterienne qu'on peut traiter avec la brosse a dents et la soie dentaire ou une solution de chlorexidine. Cette etude a donc pour objet d'etablir si la soie dentaire prealablement trempee dans la chlorexidine ameliore a elle seule la sante buccale. Methodes : Des essais paralleles randomises ont ete menes pendant trois mois chez 27 adultes qui avaient au moins 10 sites de saignement et a qui on avait demande au hasard d'utiliser de la soie dentaire trempee dans une solution de 0,12% de chlorexidine ou un placebo de sulfate de quinine. Les instructions touchant le debridement et la soie dentaire ont ete executes la premiere semaine. On a ensuite evalue la profondeur de sondage, le saignement lors du sondage, la plaque gingivale et les indices de coloration aux semaines 0, 6 et 12. La fidelite d'utilisation de la soie dentaire a ete surveillee par le biais des comptes-rendus personnels et la longueur de la soie dentaire utilisee. Resultats : Les deux groupes ont utilise tres fidelement la soie dentaire. Tous les sujets ont presente une diminution statistiquement significative de l'indice gingival (p = 0,0001). Le groupe soumis a la chlorexidine a eu une reduction statistiquement significative de la profondeur de sondage a la 6e semaine (p = 0,03); le resultat a ete plus prononce dans les sites de faible profondeur (< 4 mm au sondage), (p = 0,01) a la 6e semaine et (p = 0,01) a la 12e semaine. Le groupe soumis a la chlorexidine a aussi montre une diminution statistiquement significative du saignement au sondage chez ceux qui avaient une gingivite moderee (p = 0,01) et dans toutes les autres parties de la bouche (anterieures, p = 0,01; posterieures, p = 0,04). Les deux groupes n'ont pas souffert de facon significative des indices de la coloration et de la plaque. Conclusion : L'utilisation de la soie dentaire trempee dans la chlorexidine reduit la profondeur et le saignement au sondage chez les sujets qui ont une gingivite moderee, comparativement a la soie dentaire seule.

Key words: chlorhexidine, dental floss, bleeding on probing

INTRODUCTION

Gingivitis is an inflammatory response of the gingiva to bacteria in dental plaque. (1-3) Although gingivitis can occur on all gingival surfaces, it is more prevalent in the interproximal areas. (4) Gingivitis can be treated by mechanically removing the dental plaque by brushing teeth and dental flossing (4-6) or by chemically inhibiting plaque formation via chlorhexidine (CHX). (7-9)

Kinane et al. (1992) investigated a novel flossing device that combined the beneficial aspects of dental floss and CHX to reduce gingival bleeding in gingivitis subjects. (10) No significant differences were found between the CHX and placebo flossing devices. (10) Although the dose of CHX may have been too low, (10) another explanation is that the dental floss in the flossing device blocked the CHX from reaching the interproximal areas.

The purpose of this three-month, double-blinded, parallel randomized controlled trial (RCT) was to determine whether dental floss immersed in CHX would reduce the clinical signs of interproximal gingivitis better than a floss in placebo solution. Since rinsing with CHX is known to cause extrinsic brown tooth stain, (11) a secondary aim was to determine whether flossing with CHX would result in tooth staining.

MATERIALS AND METHODS

The study received approval from the University of British Columbia's Clinical Research Ethics Board (#C05-0513 & H05-70513) and met the requirements of the Tri-Council Policy Statement for Ethical Conduct for Research Involving Humans 1998.

Twenty-seven adults with gingivitis or localized, mild periodontitis were recruited from Vancouver, British Columbia through newspaper advertisements, community advertisements, postings on Craig's List, and referrals. The American Academy of Periodontology (1999) definitions of plaque-associated gingivitis and localized, mild chronic periodontitis were used in the study. (12) Plaque-associated gingivitis is defined as gingival inflammation, which is confined to the gingivae with no clinical attachment loss or on stable, but reduced periodontium and is only associated with dental plaque and no other local contributing factors. (12) Chronic periodontitis is defined as a slow progressive disease that results in clinical attachment loss. (12) Chronic periodontitis can be further classified by extent, which is the number of sites that are involved, and severity of clinical attachment loss. (12) For example, localized, mild periodontitis is defined as 1-2 mm of clinical attachment loss in less than 30 per cent of the total sites. (12)

Subjects were enrolled if they were non-smoking adults with gingivitis or localized, mild periodontitis. A minimum of 10 bleeding on probing (BOP) sites was required. Subjects who were accepted into the study were required to floss daily, attend all instructional sessions, and sign a consent form.

Subjects were excluded from the study if they were pregnant or planned to become pregnant within the next three months, were allergic to CHX or quinine sulfate (QS), or were required to take antibiotic premedication for dental treatment. Subjects were also excluded if they had full or partial dentures, extensive crown and bridge coverage, full orthodontic bands and brackets, or generalized, severe periodontitis (i.e., more than 30 per cent of the sites having clinical attachment loss of 5 mm or more). (12) Subjects were excluded or removed from the study if they took antibiotics, Dilantin, Cyclosporin A, Nifedipine or other calcium channel blockers, daily aspirin or anti-coagulants, CHX or whitening products.

The enrolled subjects were randomly assigned to treatment group (CHX), dental floss (Johnson & Johnson Reach[R] unflavoured waxed dental floss, Montreal, Canada) with 0.12 per cent CHX (Peridex[R] Zila Pharmaceuticals, Inc., Phoenix, Arizona), or placebo group (QS), dental floss (Johnson & Johnson Reach[R] unflavoured waxed dental floss, Montreal, Canada) with 0.1 per cent quinine sulfate solution. The placebo solution was prepared by a pharmacist to taste, smell, and to appear similar to the CHX solution. Subjects were randomized using a block design determined by a person who was not involved with the study in any other capacity. Subjects were enrolled on an ongoing basis between March 2006 and mid-September 2006, at which time the study was closed to accruement to allow the subjects to complete the 3-month study.

The study consisted of four visits over a 3-month period. All potential subjects underwent a screening visit at which medical and dental histories were recorded, periodontal condition and numbers of bleeding points were assessed, and the subject was informed about the nature of the study. If the subject met the inclusion and exclusion criteria, informed consent was obtained and the subject was scheduled for the debridement appointment.

During the debridement appointment (Week--1) calculus and plaque were removed with a combination of ultrasonic and hand instrumentation. Superficial tooth stains were removed with rubber cup prophylaxis and pumice. Flossing technique was reviewed until the subject was adept at using dental floss. Additional flossing instructions were available on a video clip on the study website and in the flossing diary. Subjects were requested to brush as usual, but refrain from using electric toothbrushes. Mouthwashes and additional professional dental hygiene services such as scaling, root planing, and rubber cup polishing were also prohibited during the study period.

Approximately one week after the debridement visit, subjects returned for baseline data collection (Week 0), which was collected in the following order: modified Loe and Silness gingival index (GI), (13) modified Lobene stain index (SI), (14) modified Silness and Loe plaque index (PI), (15) probing depths in millimeters (PD) and modified Ainamo and Bay bleeding on probing (BOP). (16)

At the end of the baseline visit, subjects received a randomly-assigned floss and flossing diary to record their flossing activity. Subjects were instructed to brush as usual then floss once a day with approximately 18" (46 cm) of dental floss. They were also requested not to rinse their mouth with water after flossing to prevent the "medicine" from being washed away.

The solution-filled floss container was placed in a heavy glass candleholder to prevent accidental spillage. Subjects were requested to ensure that the dental floss was wet at all times and were given a small bottle with extra solution to refill the floss container as needed. If the subjects thought the floss was getting dry while flossing, they were encouraged to use two pieces of dental floss, i.e., one piece for each row of teeth. All subjects received an Oral-B soft Indicator[R] toothbrush #40 (Gillette Co., Boston, Massachusetts) and Colgate[R] regular anti cavity mint toothpaste (Colgate-Palmolive Canada Inc., New York, New York) with instructions to only use these products with their assigned dental floss and not to share the study materials with family members.

At Weeks 6 and 12, measurements were retaken on the same teeth in the same order as Week 0. The dental floss, floss diary, toothbrush, and toothpaste were replenished with a new supply at Week 6. Subjects were questioned about any changes in their medical histories and whether they had experienced any side effects at each of the follow-up visits. To assess flossing compliance, the length of remaining dental floss in the container was measured and compared with the self-reported usage recorded in the flossing diary. If at the end of the study a subject presented with any one or all of these--calculus, stain, and BOP, an exit debridement was performed. All subjects were dismissed at Week 12 and requested to return to their usual oral health care professional for continuing care.

One examiner, who was blinded to the treatment assignments and calibrated before the study began, collected the clinical data on all subjects. All measurements were taken on six sites per tooth (mesial-buccal, buccal, distal-buccal, distal-lingual, lingual, and mesial-lingual) on all teeth except third molars and teeth with crown and bridge coverage. Index scores were averaged per tooth then added together and divided by number of teeth for the subject's full mouth score. Teeth were lightly dried with pressurized air prior to the measurements. Teeth were disclosed with Trace [R] disclosing solution (Young dental manufacturing company, Earth City, MO, USA) and lightly rinsed for PI. A pressure-sensitive, 3-6-9 mm periodontal probe with a point tip diameter of 0.5 mm (Kerr-Hawe Click-Probe[R] Kerr U.S.A. 1717 West Collins Avenue, Orange, CA 92867) set at 25 N (Newtons) was used to record the PD and BOP.

Statistical Analyses

An intention-to-treat protocol and whole-mouth scores were used in the statistical analyses. Whole-mouth scores for each subject were computed by adding the subject's individual tooth scores and dividing by the number of teeth. The statistical unit was the subject. According to Barbano and Clemmer (1974), subject level scores approach a continuous scale when ordinal scores are averaged to produce full-mouth scores. The continuity is further enhanced by the fact that many of these studies take the difference between a baseline reading and a reading after some subsequent treatment. (17) Cohen (2001) and Sullivan and D'Agostino (2003) also concluded, "(1) Parametric tests are sufficiently robust relative to typical violations of normality; (2) presumed statistical prohibitions against the application of parametric methods to ordinal data do not actually exist; and (3) 'ordinal' dental indices have sufficient quantitative meaning to be considered quasi-interval. For these reasons, parametric tests should not be avoided; they will be valid and usually more powerful and more easily applied to complex designs than non parametric alternatives." (18,19) Parametric tests have been used in other studies investigating the effects of CHX. (20-22) Therefore student t-tests, which are statistical tests comparing the difference between the means of two groups, (23) were used for between treatment and within treatment analyses in this study. All data were tested for normality using qq-plots, "a graphical method for diagnosing differences between the probability distribution of a statistical population from which a random sample has been taken and a comparision distribution." (24) In situations where the normality assumption appeared questionable, Wilcoxon tests were performed to ensure that the interpretations of the two methods came to the same conclusions. Alpha was set a priori at 5 per cent.

As predetermined in the research protocol, the primary outcome was BOP. GI, PI, and PD were secondary outcomes to provide additional information regarding the effects of flossing with chlorhexidine compared to flossing with placebo solution. SI was used to monitor for the common side effect known to be associated with chlorhexidine. Only predetermined stipulated hypotheses were analyzed.

Post hoc exploration of the data was done using stratification and analysis of covariance (ANCOVA) with baseline values as a covariate. Stratifications were done using baseline values. PDs were stratified into groups based on the baseline PD, but the outcome measure was in the follow-up value of millimetres. Ratios were used for the PI scores (Week 0:Week 6, Week 0:Week 12) to establish a common baseline point between the two groups.

RESULTS

Twenty-six (18 women and 8 men) of the 27 enrolled subjects completed the 12-week study. One subject withdrew at Week 6 because she was unable to "get into the flossing habit." The subject flossed for 8 days immediately after being randomized and then ceased flossing prior to the Week 6 visit. Another subject, who was on an extended holiday, missed the Week 6 visit, but continued to follow the research protocol and presented at Week 12. The subjects reported no side effects to the clinical examiner and the clinical examiner did not note any intra-oral side effects in the subjects.

At Week 0, the treatment and control groups were clinically similar for GI, PI, SI, PD, and BOP. Slight mean differences between groups were not statistically significant (Student t-test). Nevertheless, to control for the possibility of these differing baseline values on the outcomes, ANCOVA was conducted using the baseline values as a covariate. The adjusted p-values are reported in addition to the p-values from the Student t-tests.

Probing Depths (PD)

A statistically significant reduction in overall PDs was found for the subjects using the floss presoaked with CHX compared to those using the floss presoaked with the placebo solution at Week 6 (p = 0.03, adjusted p-value = 0.02) in Table 1. At Week 12, the mean overall PD for subjects using the floss presoaked in CHX remained below baseline values compared to those using the floss presoaked in QS, which rose above its baseline value; however, this was not statistically significant (p = 0.18, adjusted p-value = 0.26) as shown in Table 2.

Since dental floss is more effective in PDs that are less than 4 mm, (25-27) further analyses were conducted with the subjects' gingival sites stratified into PD < 4 mm and PD [greater than or equal to] 4 mm. At Week 6, there was a statistically significant reduction in PD in sites that were originally less than 4 mm for subjects using the CHX-soaked floss compared to those using the QS-soaked floss (p = 0.01, adjusted p-value = 0.03), but not in sites that were initially 4 mm or greater (p = 0.73, adjusted p-value = 0.50).

At Week 12, the shallow sites continued to demonstrate a statistically significant reduction in PD for the CHX group but not for the QS group (p = 0.01, adjusted p-value = 0.01). There was no statistically significant difference for PD between the CHX and QS groups for the sites that were initially 4 mm or greater (p = 0.85, adjusted p-value 0.32).

Bleeding on Probing (BOP)

A statistically significant reduction for BOP (mean change of -0.04) occurred for all subjects (p = 0.02) from Week 0 to Week 6, with smaller reductions continuing to occur up to Week 12 (mean change of - 0.02, p = 0.18). Of the initial positive bleeding sites, 83 per cent stopped bleeding in the CHX group and 78 per cent in the QS group.

As the response to CHX might be related to the level of oral health, further analyses were conducted with the subjects stratified according to "mild gingivitis" (defined for the purposes of this RCT as less than 11 initial positive BOP sites, which was the minimal number of BOP sites to be considered for inclusion into the RCT) and "moderate gingivitis" (11 or more initial positive BOP sites). Only the subjects with moderate gingivitis who used the floss presoaked in CHX had a statistically significant reduction in BOP from Week 0 to Week 6 (p = 0.01), shown in Table 3.

Since it is easier for subjects to floss the anterior teeth (canine to canine) as opposed to the posterior teeth (first premolar to second molar), (26) further analyses were conducted with the BOP sites separated into anterior and posterior areas. Statistically significant reductions in BOP occurred from Week 0 to Week 6 for the CHX group in both anterior (p = 0.01) and posterior areas (p = 0.04), seen in Table 4. In comparison, the QS group showed no statistically significant reductions in BOP from Week 0 to Week 6 (anterior, p = 0.40; posterior, p = 0.06).

From Week 0 to Week 12, subjects using the CHX-soaked floss continued to have statistically significant reductions in BOP in the anterior areas (p= 0.01). All other comparisons for BOP between the CHX and QS groups were not statistically significant, depicted in Table 5.

Plaque Index (PI)

Over the 12-week study, a constant mean PI was found for subjects using the CHX-soaked floss and this appeared to differ from the increasing PI in patients enrolled in the QS group shown in Figure 1. However, there was no statistically significant difference between these groups.

[FIGURE 1 OMITTED]

Gingival Index (GI)

All subjects had statistically significant reductions in mean GI scores from Week 0 to Week 6 (mean change of -0.56, p < 0.001) as well as from Week 0 to Week 12 (mean change of-0.58, p < 0.0001) demonstrated in Figure 2.

[FIGURE 2 OMITTED]

Stain Index (SI)

There was no statistically significant difference between the CHX and QS groups for stain at Week 6 (p = 0.91, adjusted p = 0.52) or Week 12 (p = 0.18, adjusted p-value = 0.32). Both groups had a slight, but not statistically significant, increase in stain over the 12 weeks (mean change 0.05, p = 0.14).

Flossing Compliance

Both groups demonstrated high levels of flossing compliance with no statistically significant differences between the groups. At Week 6, the self-reported median flossing compliance was 98 per cent for the CHX group and 97 per cent for the QS group and at Week 12 it was 100 and 93 per cent respectively. Median yards of floss used ranged from 35 to 43 yards (about 32-39 m) per 6-week period.

DISCUSSION

The introduction of a daily flossing regimen resulted in an overall benefit for all study subjects. Flossing, as shown by the results of the positive control group, resulted in statistically significant reductions in BOP scores from Week 0 to Week 6, and to a lesser degree up to Week 12. All subjects also had statistically significant reductions in GI scores over the 12-week study. The reductions in bleeding and gingival index scores found in this RCT are similar to the results found in other studies, which have demonstrated the beneficial effects of flossing for the treatment of gingivitis. (25-28)

However, presoaking the dental floss in CHX solution had additional benefits compared to the floss soaked in the placebo solution. The CHX-soaked dental floss had statistically significant reductions for probing depths in sites that were initially less than 4 mm compared to the floss in placebo solution, which did not demonstrate any statistically significant PD reductions. Both groups did not have statistically significant reductions in PDs for sites that were initially 4 mm or more most likely because dental floss can only effectively deplaque sulcular depths to a maximum of 3 mm. (25,29,30) The data suggests that the CHX-soaked floss may have been able to carry the CHX into the interproximal area to produce a reduction in PDs less than 4 mm similar to the effects seen by oral irrigation with CHX solution. For example, Flemmig et al. (1990) demonstrated a reduction in probing depths (mean reduction of 4.6 per cent at 6 months, p < 0.05) in shallow sulci by irrigating with 0.06 per cent CHX rinse. (31) Although the method of applying CHX differs, oral irrigation may flush CHX subgingivially into the sulcus (31) just as dental floss may carry CHX into the sulcus to reduce probing depths.

The CHX-soaked floss also demonstrated additional BOP reductions for subjects with 11 or more initial BOP sites compared to the QS-soaked floss. The subgroup of moderate gingivitis subjects using the CHX-soaked floss had a statistically significant reduction in BOP from Week 0 to Week 6, which continued to a lesser degree up to Week 12. CHX mouth rinse has been shown in other studies to reduce bleeding, with reductions ranging from 46-67%. (8,9,27,31-35) However, according to Cumming and Loe (1973) and Caton et al. (1993) CHX mouth rinses may have limited effects interproximally. (4,36) In this study, the CHX was applied interproximally via a presoaked dental floss and was able to exert an additional effect over what was achieved with flossing alone.

Since individuals can floss the anterior teeth (canine to canine) more effectively than the posterior teeth (first premolar to second molar), (26) analyses were conducted with the BOP sites grouped into anterior and posterior sites. The subjects using the CHX-soaked floss had statistically significant reductions in BOP over the 12 weeks for the anterior areas, but only had statistically significant reductions in the posterior areas up to Week 6. The results of this study are similar to those of Wong and Wade (1985) in that the subjects were able to floss the anterior teeth more effectively than the posterior teeth. (26)

The Silness and Loe (1964) plaque index requires an examiner to see the amount and location of plaque in order to assign a score for the tooth (15); it is the easiest, most portable and cost-effective method to use in the field. (37) However, it is not possible to visualize the interproximal surfaces, which were targeted by the CHX-soaked floss. A more sensitive measure that quantifies plaque in the interproximal area needs to be developed.

Numerous studies have shown CHX to be an effective anti-plaque agent. (11) In this RCT, the researchers were unable to assess the anti-plaque effects of the CHX-soaked floss because of the limitations of the plaque index; however, the statistically significant reductions in PDs and BOP indicate that the CHX-soaked floss was having a beneficial effect in the interproximal area compared to the placebo-soaked floss in similar sites.

The other benefit of using a CHX-soaked dental floss rather than a CHX mouth rinse is that the floss method may minimize tooth staining. CHX mouth rinse is known to cause tooth staining within a few days of use in 3 out of 4 individuals who use it (11) and this is the primary reason for low compliance with the CHX mouth rinse regimen. (11,21,38-42) However, in this RCT there was no noticeable tooth staining in subjects using the CHX-soaked floss. Both the CHX and QS groups had slight increases in tooth stain over the 12 weeks but this was not statistically significant and may be attributed to dietary sources such as tea and coffee drinking.

Subjects' compliance with the flossing regime was excellent, with most subjects flossing daily. Although the subjects used more than twice the amount of floss than was expected, the high usage corresponded to the high numbers of self-reported flossing days, indicating that compliance was high.

CONCLUSION

In this efficacy study, dental flossing alone reduced GI and BOP scores and therefore, is an effective method for treating gingivitis. However, dental floss presoaked in a 0.12 per cent CHX solution offers additional benefits for the treatment of gingivitis such as, reducing PDs in shallow sulcular sites (PD < 4 mm) and bleeding in subjects with moderate amounts of gingival bleeding. The CHX-soaked floss is also more effective for reducing bleeding in all areas of the mouth, but more so in the anterior sites, than the floss soaked in the placebo solution. Although it was not possible to discern a difference between the groups for interproximal plaque levels because of the limitations of the PI, the other positive results associated with using the CHX-soaked floss indicates that flossing with the CHX-soaked floss provides additional beneficial effects in the interproximal area compared to flossing alone.

ACKNOWLEDGEMENTS

The authors would like to thank Dr. Ian Low for the use of his dental clinic for the study and Dr. Ryan Woods, Senior Statistician of the Canadian HIV Trials Network, for conducting and advising us on the statistical analyses. This study was generously supported by grants from the CFDHRE and BCDHA.

REFERENCES

1. Mariotti A. Dental plaque-induced gingival diseases. Ann Periodontol. 1999;4:7-17.

2. Tolle SL. Periodontal and risk assessment. In: Darby ML, Walsh MM. Dental hygiene theory and practice. 2nd ed. Missouri:WB Saunders; 2003. 272-311.

3. Loe H, Theilade E, Jensen SB. Experimental gingivitis in man. J Periodontol. 1965;36:177-87.

4. Caton JG, Blieden TM, Lowenguth RA, Frantz BJ, Wagener CJ, Doblin JM, Stein SH, Proskin HM. Comparison between mechanical cleaning and an antimicrobial rinse for the treatment and prevention of interdental gingivitis. J Clin Periodontol. 1993;20:172-8.

5. Gjermo P, Flotra L. The effect of different methods of interdental cleaning. J Periodontal Res. 1970;5:230-6.

6. Bergenholtz A. Mechanical cleaning in oral hygiene. In: Frandsen A, editor. Oral Hygiene. Copenhagen: Munksgaard; 1972. 27-62.

7. Lang NP, Brexc MC. Chlorhexidine digluconate- an agent for chemical plaque removal and prevention of gingival inflammation. J Periodontal Res. 1986;21:74-89.

8. Charles CH, Mostler KM, Bartels LL, Mankodi SM. Comparative antiplaque and antigingivitis effectiveness of a chlorhexidine and an essential oil mouth rinse: 6-month clinical trial. J Clin Periodontol. 2004;31:878-84.

9. Grossman E, Reiter G, Sturzenberger OP, De La Rosa M, Dickinson TD, Ferretti GA, Ludlam GE, Meckel AH. Six-month study of the effects of a chlorhexidine mouth rinse on gingivitis in adults. J Periodontal Res. 1986;21:33-43.

10. Kinane DF, Jenkins WMM, Paterson AJ. Comparative efficacy of the standard flossing procedure and a new floss applicator in reducing interproximal bleeding: a short-term study. J Periodontol. 1992;63:757-60.

11. Imai PH. A review of the different methods of applying chlorhexidine in the oral cavity. Can J Dent Hygiene. 2006;40:69-79.

12. Armitage GC. Development of a classification system for periodontal diseases and conditions. Ann Periodontol. 1999;4:1-6.

13. Loe H, Silness J. Periodontal disease in pregnancy. I. Prevalence and severity. Acta Odont Scand. 1963;21:533-51.

14. Lobene RR. Effect of dentifrices on tooth stains with controlled brushing. JADA. 1968;77:849-55.

15. Silness J, Loe H. Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal condition. Acta Odont Scand. 1964;22:121-35.

16. Ainamo J, Bay I. Problems and proposals for recording gingivitis and plaque. Int Dent J. 1975;25:229-35.

17. Barbano JP, Clemmer BA. A comparison of analyses of dicohotmous and severity data in clinical trials using dental data. J Periodontal Res. 1974;9 (Suppl 14):129-42.

18. Cohen ME. Analysis of ordinal dental data: evaluation of conflicting recommendations. J Dent Res. 2001;80(1):309-13.

19. Sullivan LM, D'Agostino RB Sr. Robustness and power of analysis of covariance applied to ordinal scaled data as arising in randomized controlled trials. Stat Med. 2003;22(8):1317-34.

20. Santos S, Herrera D, Lopez E, O'Connor A, Gonzalez I, Sanz M. A randomized clinical trial on the short-term clinical and microbiological effects of the adjunctive use of a 0.05% chlorhexidine mouth rinse for patients in supportive periodontal care. J Clin Periodontol. 2004;31:45-51.

21. Pannuti CM, Saraiva C, Ferraro A, Falsi D, Cai S, Lotufo RFM. Efficacy of a 0.05% chlorhexidine gel on the control of gingivitis in Brazilian mentally handicapped patients. J Clin Periodontol. 2003;30:573-6.

22. Sanz M, Vallcorba N, Fabregues S, Muller I, Herkstroter F. The effect of a dentifrice containing chlorhexidine and zinc on plaque, gingivitis, calculus and tooth staining. J Clin Periodontol. 1994;21:431-37.

23. Trochim WMK. The research methods knowledge base. 2nd ed. Cincinnati: Atomic Dog Publishing; 2001;352.

24. QQ plot. Wikipedia, the free encyclopedia. [Online]. 2007 [Cited 2007 Nov 21]; Available from: URL: http://en.wikipedia.org/wiki/Q-Q_plot Cited November 21, 2007.

25. Asadoorian J. Flossing: Canadian Dental Hygienists Association position paper. Can J Dent Hygiene. 2006;40:112-25.

26. Wong CH, Wade AB. A comparative study of effectiveness in plaque removal by Superfloss[R] and waxed dental floss. J Clin Periodontol. 1985;12:788-95.

27. Graves RC, Disney JA, Stamm JW. Comparative effectiveness of flossing and brushing in reducing interproximal bleeding. J Periodontol. 1989;60:243-7.

28. Carr MP, Rice GL, Horton JE. Evaluation of floss types for interproximal plaque removal. Am J Dent. 2000;13:212-4.

29. Waerhaug J. Effect of toothbrushing on subgingival plaque formation. J Periodontol. 1981;52:30-4.

30. Bellamy P, Barlow A, Puri G, Wright KL, Mussett A, Zhou X. A new in vivo interdental sampling method comparing a daily flossing regime versus a manual brush control. J Clin Dent. 2004;15:59-65.

31. Flemmig TF, Newman MG, Doherty FM, Grossman E, Meckel AH, Bakdash MB. Supragingival irrigation with 0.06% chlorhexidine in naturally occurring gingivitis. I. 6 month clinical observations. J Periodontol. 1990;61:112-7.

32. Joyston-Bechal S, Hernaman N. The effect of a mouth rinse containing chlorhexidine and fluoride on plaque and gingival bleeding. J Clin Periodontol. 1993;20:49-53.

33. Grindemann LJMM, Timmerman MF, Ijserman Y, Van der Velden U, Van der Weijden GA. Stain, plaque and gingivitis reduction by combining chlorhexidine and peroxyborate. J Clin Periodontol. 2000;27:9-15.

34. Hammerle CHF, Joss A, Lang NP Short-term effects of initial periodontal therapy (hygienic phase). J Clin Periodontol. 1991;18:233-9.

35. Haffajee AD, Cugini MA, Dibart S, Smith C, Kent Jr. RL, Socransky SS. The effect of SRP on the clinical and microbiological parameters of periodontal diseases. J Clin Periodontol. 1997;24:324-34.

36. Cumming BR, Loe H. Optimal dosage and method of delivering chlorhexidine solutions for the inhibition of dental plaque. J Periodontal Res. 1973;8:57-62.

37. Fischman SL. Current status of indices of plaque. J Clin Periodontol. 1986;13:371-4.

38. Flotra L, Gjermo P, Rolla G, Waerhaug J. Side effects of chlorhexidine mouth washes. Scand J Dent Res. 1971;79:119-25.

39. Overholser CD Jr. Longitudinal clinical studies with antimicrobial mouthrinses. J Clin Periodontol. 1988;15:517-9.

40. Loe H, Schiott CR, Glavind L, Karring T. Two years oral use of chlorhexidine in man. J Periodontal Res. 1976;11:135-44.

41. Carpenter GH, Pramanik R, Proctor GB. An in vitro model of chlorhexidine-induced tooth staining. J Periodontal Res. 2005;40:225-30.

42. Eriksen HM, Nordbo H, Kantanen H, Ellingsen JE. Chemical plaque control and extrinsic tooth discoloration: a review of possible mechanisms. J Clin Periodontol. 1985;12:345-50.

Pauline H. Imai, DIPDH, BDSC, MSC; Edward E. Putnins, DMD, DIPPERIO, MRCD (C), MSC, PHD; Donald M. Brunette, BSC, MSC, PHD

From the University of British Columbia, Vancouver

Submitted 30 Apr. 2007; Revised 2 Sept. 2007; Accepted 5 Nov. 2007

Correspondence to: DM Brunette, Oral Biological and Medical Sciences, University of British Columbia, 2199 Westbrook Mall, Vancouver, BC V6T 1Z3; [e-mail]

Initial probing)                 CHX group (n = 12)  QS group (n = 14)
depth sites (PD)                 Mean  SD            Mean  SD

Week 0
Overall PD                       2.31  0.34          2.42  0.32
  PD < 4mm                       2.11  0.05          2.24  0.03
  PD [greater than or equal to]  4.07  0.03          4.14  0.04
  4mm

Week 6
Overall PD                       2.20  0.24          2.40  0.19
  PD < 4mm                       2.06  0.12          2.23  0.16
  PD [greater than or equal to]  3.73  1.18          4.05  0.09
  4 mm

                                                     Adjusted p-value
                                 p-value (2 sample   (ANCOVA with
Initial probing                  t-test or Wilcoxon  baseline as
depth sites (PD)                 Rank Sum test)      covariate)

Week 0
Overall PD                       0.19
  PD < 4mm                       0.06
  PD [greater than or equal to]  0.14
  4mm

Week 6
Overall PD                       0.03                0.02
  PD < 4mm                       0.01                0.03
  PD [greater than or equal to]  0.73                0.50
  4 mm

Table 1: Comparison of overall probing depths (PD) and stratified
probing depths (PD < or [greater than or equal to] 4 mm) for
chlorhexidine (CHX) and placebo (QS) groups at Weeks 0 and 6.

                                      CHX group   QS group
Initial probing                       (n = 12)    (n = 14)
depth (PD)                            Mean  SD    Mean  SD

Week 0
Overall PD                            2.31  0.34  2.42  0.32
  PD < 4mm                            2.11  0.05  2.24  0.03
  PD [greater than or equal to] 4mm   4.07  0.03  4.14  0.04

Week 12
Overall PD                            2.29  0.35  2.44  0.20
  PD < 4mm                            2.09  0.16  2.26  0.11
  PD [greater than or equal to] 4 mm  3.76  1.13  4.04  0.06

                                      p-value (2      Adjusted p-value
                                      sample t-test   (ANCOVA with
Initial probing                       or Wilcoxon     baseline as
depth sites (PD)                      Rank Sum test)  covariate)

Week 0
Overall PD                            0.19
  PD < 4mm                            0.06
  PD [greater than or equal to] 4mm   0.14

Week 12
Overall PD                            0.18            0.26
  PD < 4mm                            0.01            0.01
  PD [greater than or equal to] 4 mm  0.85            0.32

Table 2: Comparison of overall probing depths (PD) and stratified
probing depths (PD < or [greater than or equal to] 4 mm) for
chlorhexidine (CHX) and placebo (QS) groups at Weeks 0 and 12.

                                             Mean change
Gingivitis severity                Floss     from Week 0
(initial BOP sites)                used   N  to Week 6    SD

Moderate                           CHX    8  -0.12        0.09
  ([greater than or equal to] 11)
Moderate                           QS     6  -0.08        0.06
  ([greater than or equal to] 11)
Mild (<11)                         CHX    4   0.02        0.04
Mild (<11)                         QS     8   0.02        0.06

                                   p-value
                                   (Wilcoxon
Gingivitis severity                Signed Rank
(initial BOP sites)                test)

Moderate                           0.01
  ([greater than or equal to] 11)
Moderate                           0.06
  ([greater than or equal to] 11)
Mild (<11)                         0.50
Mild (<11)                         0.73

Table 3: Comparison of mean change in bleeding on probing (BOP) from
Week 0 to Week 6 for subjects stratified according to mild gingivitis
(< 11 initial BOP sites) and moderate gingivitis
([greater than or equal to] 11 initial BOP sites) using floss soaked in
either chlorhexidine (CHX) or placebo (QS).

Area in subjects'     Floss                           P-value
mouth                 used   N   Mean change  SD      (Paired t-tests)

Anterior (canine to   CHX    12  -0.02        0.001   0.01
  canine)             QS     13   0.00        0.0004  0.40
Posterior (bicuspids  CHX    12  -0.02        0.002   0.04
  to molars)          QS     13  -0.01        0.0003  0.06

Table 4: Comparison of the mean changes in bleeding on probing (BOP) for
sites stratified according to anterior or posterior areas of the mouth
in subjects using either the floss soaked in chlorhexidine (CHX) or
placebo from Week 0 to Week 6.

Area in subjects'                                     P-value
mouth                 Floss  N   Mean change  SD      (Paired t-tests)

Anterior (canine to   CHX    13  -0.02        0.001   0.01
  canine)             QS     14  -0.01        0.0004  0.09
Posterior (bicuspids  CHX    13  -0.01        0.001   0.23
  to molars)          QS     14   0.00        0.0003  0.32

Table 5: Comparison of the mean changes in bleeding on probing (BOP) for
sites stratified according to anterior or posterior areas of the mouth
in subjects using either the floss soaked in chlorhexidine (CHX) or
placebo from Week 0 to Week 12.