|
|
||||||||
|
J Am Dent Assoc, Vol 137, No 12, 1649-1657.
© 2006 American Dental Association | ![]() |
COVER STORY |
| ABSTRACT |
|---|
|
|
|---|
Results. Seventeen studies support the antiplaque, antigingivitis effects of dentifrices containing 0.30 percent triclosan, 2.0 percent Gantrez copolymer. There was no evidence of efficacy for triclosan products containing either soluble pyrophosphate or zinc citrate. Dentifrices with stannous fluoride had statistically significant, but marginally clinically significant, evidence of an antiplaque effect; however, there was both a statistically and clinically significant antigingivitis effect. The largest body of studies (21 studies) supported the efficacy of mouthrinses with essential oils. A smaller body of studies (seven) supported a strong antiplaque, antigingivitis effect of mouthrinses with 0.12 percent chlorhexidine. Results for mouthrinses with cetylpyridinium chloride varied and depended on the products formula.
Conclusions. The studies in this systematic review provide strong evidence of the antiplaque, antigingivitis effects of multiple agents. These results support the use of these agents as part of a typical oral hygiene regimen.
Key Words: Antiplaque; antigingivitis; systematic review; plaque control
Multiple antigingivitis and antiplaque over-the-counter products are available commercially for patients. These products are primarily in the form of a dentifrice or a mouthrinse, and the active agents involved include triclosan (dentifrice), stannous fluoride (dentifrice), a combination of essential oils (mouthrinse) and cetylpyridinium chloride (CPC) (mouthrinse). In addition, in the United States, a mouthrinse containing chlorhexidine is available as a prescription drug.
The proper formulation of these active agents into dentifrices and/or mouthrinses is extremely important to maintain the bioavailability of the agents and, in some cases, to improve their substantivity. Thus, different formulations of the same active agents may have different levels of efficacy. This increases the number of choices in the marketplace for products containing these agents and makes it more difficult to evaluate their efficacy.
The clinical evaluation of these products includes short-term trials ranging from four days to two months and long-term clinical trials, most of which are six months in length. The four-day trials are used primarily to evaluate the antiplaque effect of these products.15 The intermediate-length trials (two weeks to two months) can evaluate both the antiplaque and antigingivitis efficacy of these products. As demonstrated by the three-week, no-oral-hygiene model of Löe and colleagues,6 most people will develop gingivitis in this time frame. The limitation of these intermediate-length trials is that researchers cannot investigate the long-term efficacy of the product, which would reflect more accurately the patients actual use of the product. For this reason, six-month trials have been used to evaluate efficacy by the U.S. Food and Drug Administration (FDA) in its approval of 0.12 percent chlorhexidine as a prescription drug in the United States, as well as by the American Dental Association in its Seal of Acceptance Program.7,8
Although a variety of antiplaque, antigingivitis agents have been evaluated in randomized, six-month trials, no comprehensive systematic reviews of these studies, to my knowledge, have been conducted. The only systematic review of six-month clinical trials of antiplaque, antigingivitis agents has been a review of a triclosan dentifrice containing a copolymer.9
Therefore, the goal of this systematic review was to evaluate the efficacy of antiplaque, antigingivitis agents in studies involving six-month randomized clinical trials.
MEDLINE search.
Initially, a dental hygienist and I used MEDLINE to search for the studies. We used the abstracts to eliminate any studies that were not relevant. We then obtained the full-length articles of all relevant studies. We used the following sets of key words, for which the resulting number of studies are shown parenthetically:
After duplicate studies were removed, the above search resulted in a total of 838 unique articles. We reviewed the titles to determine if the studies were clinical trials of six months in duration and evaluated plaque and gingivitis.
We conducted a hand search from the initial findings to locate any studies that may have been missed. We also contacted manufacturers to determine if they had conducted any unpublished studies.
The inclusion criteria used to screen published and unpublished reports were as follows:
I added this criterion because only three studies did not use this index, but instead used the plaque index (PI).11 Studies that used the PI to evaluate plaque were included in the evaluation of gingival inflammation.
To be included in the evaluation of gingival inflammation the study had to use the gingival index (GI)12 or the modified gingival index (MGI).13 There were too few studies that used the same bleeding index (BI) to use that index as an outcome measure. However, if the study used only a BI, it still was included in the plaque evaluation as long as the investigators used the Turesky modification of the Quigley-Hein Index to evaluate plaque.
the sample size of the study groups had to be specified.
We found a total of 80 full-length published and unpublished reports that evaluated the antiplaque and antigingivitis effects of mouthrinses and dentifrices in six-month trials. This included 53 published reports and 27 unpublished research reports. Nine of the studies included both a published report and an unpublished research report. Three of the reports from manufacturers had significant problems with regard to protocol violations (in one case, the examiner became ill in the middle of the study) and/or recorded only partial mouth measurements; I excluded these studies from the meta-analysis. An additional 27 reports were eliminated because they evaluated agents that are no longer being manufactured and/or evaluated agents about which there were too few studies to analyze.
We found a total of 50 articles and/or reports that met the inclusion criteria.1451 Because some of these studies compared multiple products with placebo groups, there were a total of 70 active groups in these studies. Two reports were unpublished reports that were identified to the plaque subcommittee formed by the FDA to evaluate over-the-counter agents, 13 were unpublished reports that were provided by manufacturers and 36 were published articles (nine of these studies also had unpublished research reports). We abstracted the studies that met these criteria for outcome variables (PI, GI), sample size, blinding, randomization, agent types and formulations, and types of controls.
Data analysis.
I performed separate analyses of efficacy for each of the active agents. I first evaluated the data for heterogeneity (a statistical test for consistency among study results). A random-effects model was used to evaluate the overall efficacy of the data52,53 when heterogeneity was present. I also noted statistically significant heterogeneity when present. Because unpublished research reports were included in the analysis, I compared published and unpublished results. However, I found a sufficient number of published and unpublished reports only for the evaluation of essential oils to enable me to make this comparison. I noted no differences in the results between the published and unpublished studies.
As noted above, some studies had multiple active arms of either similar or dissimilar agents. Data for the arms were entered and analyzed as though they were separate studies. I analyzed the data both with and without these multiple-arm studies, and the findings remained unchanged.
Stannous fluoride.
The dentifrices containing stannous fluoride (four published studies2124 and one unpublished study [M.E. Mallatt and colleagues, 2005, unpublished data]) exhibited a statistically significant, but small antiplaque effect (mean standardized difference [Std. Diff.] between groups = 0.168). (A mean standardized difference is a mean difference in the active agents effect minus the control agents effect adjusted by the variability of each study. This standardization accounts for the difference in variability among multiple studies.)
Triclosan/copolymer.
In contrast, the dentifrice composed of 0.30 percent triclosan, 2.0 percent Gantrez copolymer exhibited significant results for 14 of the 18 arms (representing 17 studies) and a substantially larger effect (Std. Diff. = 0.823). Because there was statistically significant heterogeneity, I used the random effects model to evaluate the efficacy of the agent. The overall analysis of the efficacy of the triclosan/copolymer agent using a random-effects model resulted in a highly significant (P < .0001) mean group difference that favored the active agent.
Triclosan/soluble pyrophosphate.
Although studies of dentifrices containing the triclosan/soluble pyrophosphate agent resulted in marginally statistically significant results, the test for heterogeneity also was significant. The studies were inconsistent, however, because three of the four resulted in nonsignificant results and, thus, did not provide sufficient evidence that this is an effective antiplaque agent.
Triclosan/zinc citrate.
The triclosan/zinc citrate agent also resulted in a nonsignificant result.
Of the dentifrices evaluated, then, only those containing the triclosan-copolymer agent showed both a clinically significant and a statistically significant antiplaque effect.
Antiplaque effect of mouthrinses.
In this meta-analysis, I evaluated the antiplaque efficacy of three active agents in mouthrinses: 0.12 percent chlorhexidine, CPC and essential oils.
Chlorhexidine.
The 0.12 percent chlorhexidine mouthrinse had a consistent antiplaque effect (Std. Diff. = 1.040), and the results of all studies were statistically significant.
CPC.
In the efficacy analysis of CPC, four1416 (V. Segreto and E. Collins, 1993, unpublished data) of the seven studies exhibited statistical significance and three did not (S.G. Ciancio, 1979, unpublished data; R.R. Lobene, 1977, unpublished data; and Menaker and colleagues, 1986, unpublished data). The test for heterogeneity was statistically significant. I should point out that the products evaluated in the studies also varied (Cepacol Antibacterial Mouthwash, Combe, White Plains, N.Y., containing 0.05 percent CPC [Ciancio, 1979, unpublished data; Lobene, unpublished data, 1977]; Scope Mouthwash, Procter & Gamble, Cincinnati, containing 0.045 percent CPC [L. Menaker and colleagues, 1986, unpublished data]; a 0.05 percent CPCcontaining mouthrinse14 [Segreto and Collins, 1993, unpublished data]; and two mouthrinses containing 0.07 percent CPC, one with an alcohol vehicle15 and one without alcohol16). Thus, there was a great deal of heterogeneity in both the CPC agents evaluated and in the results obtained, with some of the agents exhibiting antiplaque effects and some not exhibiting these effects.
Essential oils.
The majority of the studies (20 studies, 25 arms) evaluated mouthrinses containing essential oils. Of these studies, only one failed to show statistical significance. Although the test for heterogeneity was positive, the study results clearly support the antiplaque efficacy of the essential oils (Std. Diff. = 0.852, P < .0001).
Antigingivitis effects.
In the analysis of gingivitis, I found more variety with regard to the index used than I did with regard to the index used to evaluate plaque. Two studies reported using only a BI, and a few used a BI and a GI. However, the number of studies that used a BI was insufficient for analysis. The remaining studies used either the GI21 or the MGI11 to evaluate gingivitis. Table 1
![]()
MATERIALS AND METHODS
TOP
ABSTRACT
MATERIALS AND METHODS
RESULTS
DISCUSSION
CONCLUSION
REFERENCES
The focused question for this systematic review was as follows: Are mouthrinses or dentifrices effective (and which ones) as antiplaque and/or antigingivitis agents in six-month randomized clinical trials of adults 18 years and older?
![]()
RESULTS
TOP
ABSTRACT
MATERIALS AND METHODS
RESULTS
DISCUSSION
CONCLUSION
REFERENCES
Antiplaque effect of dentifrices.
Forty-seven (94 percent) of the 50 studies used the Turesky modification of the Quigley-Hein Index, and I limited my analysis of the antiplaque efficacy to these studies. (The other three studies used the PI.) Owing to the large number of studies that had plaque data, I divided the forest plots presenting the plaque results into studies that evaluated dentifrices (29 active arms) and studies that evaluated mouthrinses (38 active arms) (data available online). (Forest plots are a graphical representation of the effect of an active agent over a control agent for all of the studies included in a systematic review.)
presents the results for all agents that were evaluated with the GI (33 studies, 45 active arms), and Table 2
(page 1654) presents the results for agents evaluated with the MGI (17 studies, 22 active arms).
|
|
Significant heterogeneity was present in the analysis of antigingivitis effects for 0.12 percent chlorhexidine (P = .013), essential oils (P < .001), stannous fluoride dentifrice (P = .010), CPC (P = .004) and triclosan with 2.0 percent Gantrez copolymer (P < .001). I found the most consistent results for the 0.12 percent chlorhexidine mouthrinse (five of the six active arms demonstrated statistically significant effects) and for the triclosan with 2.0 percent Gantrez copolymer dentifrice (12 of the 16 arms demonstrated statistically significant effects). Thus, the evidence was consistent and strong in support of the antigingivitis effects of these agents.
The analysis showed moderately consistent results with regard to the antigingivitis effects for mouthrinses containing essential oils and for stannous fluoride dentifrices; the results of one-half of the studies were statistically significant for each agent, but all of the studies demonstrated antigingivitis effects. Similar to the evaluation of the antiplaque effects of mouthrinses containing CPC, I found both statistical heterogeneity and a variety of formulations evaluated; thus, it was difficult to reach conclusions about this agent, although the results of studies of individual CPC products were similar to those for other types of active agents.
Researchers used the MGI primarily in studies evaluating essential oilcontaining mouthrinses (Table 2
). Thirteen of the 17 studies1820,30,45,48 (S. Mankodi and colleagues, 1993, two arms, unpublished data; S. Mankodi and colleagues, 1989, two arms, unpublished data; N. Sharma and colleagues, two arms, 1997, unpublished data; K. Bauroth and colleagues, 2004, unpublished data) had statistically significant results with regard to the antigingivitis properties of essential oilcontaining products compared with the control agents. Again, there was significant heterogeneity (P < .0001) for the comparison. However, all of the studies favored the essential oilcontaining mouthrinse over the control mouthrinse (Std. Diff. = 0.762). Thus, this evaluation by the MGI provides strong evidence in support of the antigingivitis properties of essential oilcontaining mouthrinses.
Few studies evaluated the relative efficacy of the different agents described above. Four studies42,43,45 (Segreto and Collins, 1993, unpublished data) compared mouthrinses containing essential oils with chlorhexidine mouthrinses. The studies compared active agents with control agents, as well as with each other. The results showed statistically significant antiplaque effects for both agents in all four studies. In all cases, 0.12 percent chlorhexidine exhibited greater antiplaque effects than did mouthrinses containing essential oils (the effect for essential oils was about 60 percent of that for the 0.12 percent chlorhexidine mouthrinse).
The results were similar for gingivitis. The four studies showed a statistically significant advantage for the active agents over the control agents. Mouthrinses containing essential oils had about 60 percent of the antigingivitis effect of 0.12 percent chlorhexidine; however, the difference in antigingivitis effects of the agents was close to, but failed to reach, statistical significance (P = .068). There were too few head-to-head comparisons of the other active agents (two or fewer) to evaluate.
| DISCUSSION |
|---|
|
|
|---|
The largest number of studies of mouthrinses evaluated essential oils (22 studies, four with two active arms). These studies show clearly that this agent is effective as both an antiplaque and an antigingivitis agent. Six studies evaluated the 0.12 percent chlorhexidine mouthrinse, and the results of these were remarkably consistent. The results were inconsistent for products containing CPC; however, the concentrations of CPC varied from 4.5 to 7 percent. In addition, the two studies that evaluated the 7 percent concentration involved formulations that were alcohol-based15 or nonalcohol-based.16 While the six-month results were promising for the nonalcohol-based agent, more long-term studies are needed to provide the same level of evidence that exists for the other agents.
The goal of antiplaque, antigingivitis agents is to decrease gingival inflammation so that destructive periodontal disease will not develop. The evidence demonstrates clearly that mouthrinses containing 0.12 percent chlorhexidine or essential oils and dentifrices containing triclosan with 2.0 percent Gantrez copolymer or stannous fluoride reduce the level of gingival inflammation. It is not clear, however, what level of reduction is necessary to decrease or prevent periodontal disease. However, gingival inflammation is a necessary, but insufficient, condition for the initiation and progression of periodontal disease. Therefore, reducing gingival inflammation with these agents is highly desirable. More studies are needed to determine if the level of reduction of gingival inflammation produced by these agents is sufficient to prevent, or slow the progress of, periodontal disease.
Dentifrices with stannous fluoride demonstrated a statistically significant antiplaque effect, but most likely not a clinically significant effect, because the effect was much smaller than that of the other agents. These results suggest that the main mechanism of action in stannous fluoride agents is not in the suppression of plaque mass, but in altering the ability of the plaque to affect the levels of gingivitis. Bacca and colleagues17 suggested that the efficacy of stannous fluoride in reducing gingivitis is due to its alteration of the virulence and effects of the plaque composition, not to the overall quantity of plaque.
Recent studies have evaluated chemotherapeutic agents in relationship to mechanical plaque control. The researchers compared the efficacy of an essential oilcontaining mouthrinse with that of flossing.1820 Two of the studies18,20 demonstrated that the chemotherapeutic control of interproximal plaque and gingivitis by an essential oilcontaining mouthrinse met or exceeded the interproximal control of flossing. The third study19 demonstrated that an essential oilcontaining mouthrinse adds to the interproximal control of plaque and gingivitis achieved with flossing alone. In concert with the results of this review, there is strong evidence1820 of the benefit of using chemotherapeutic agents in addition to mechanical methods of brushing and flossing in adults to control plaque and gingivitis. Adding one chemotherapeutic agent to the typical oral hygiene regimen will reduce the level of gingival inflammation in these patients.1820 The most likely benefit is the prevention and/or reduction of periodontal disease, but further studies are needed to demonstrate that use of these agents will result in a lower prevalence and severity of periodontal disease.
One of the goals in reporting a series of studies that evaluate the efficacy of an agent is to determine the factors that influence study outcomes. I evaluated the studies in this meta-analysis to determine which factors might account for the difference between placebo effects and active agent effects. I evaluated the following factors: baseline plaque and gingivitis levels and the level of supervision (that is, whether subjects were supervised on a daily basis in their use of the assigned product) provided for the study agents. I could find no relationship between baseline plaque and/or gingivitis levels and the efficacy of the agents.
The only factor that influenced the outcomes of these studies was the use of different gingival indexes. In evaluating the efficacy of essential oilcontaining mouthrinses (the data were insufficient for the other agents to make this comparison), I found that the MGI resulted in a larger difference between the effects of the placebo and those of the active agents than did the GI. The MGI has a wider scale (04) than the GI (03). It appears that the additional category helps in the identification of antigingivitis effects. These results suggest that supervision in the use of these agents is not necessary in these types of studies, but use of the MGI will improve researchers ability to show differences between active and placebo agents.
| CONCLUSION |
|---|
|
|
|---|
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. R. Ashford and M. Skelley Oral Care and the Elderly Swallowing and Swallowing Disorders (Dysphagia) , March 1, 2008; 17(1): 19 - 26. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |