The Antiplaque/Anticariogenic Efficacy of Salvadora Persica (Miswak) Mouthrinse in Comparison to That of Chlorhexidine

A Systematic Review and Meta-analysis

Elaf Jassoma; Lina Baeesa; Heba Sabbagh

Disclosures

BMC Oral Health. 2019;19(64) 

In This Article

Results

Systematic Review Results

This revised search strategy yielded a total of 1135 potentially eligible articles that were published between January 1970 and December 2018. The results generated across the five databases were as following: MEDLINE-PubMed (77), Cochrane Central Register of Controlled Trials (28), Wiley Online Library (193), ScienceDirect (164), and Google Scholar (674). From these, 123 duplicates were excluded. The titles of the remaining 1012 articles were evaluated. Subsequently, 935 articles were eliminated after the titles and abstracts had been assessed. The full texts of the remaining 77 papers were reviewed and compared to the inclusion criteria chosen.

Fifty-nine articles were excluded for the following reasons: because they were not in-vivo studies (four articles), did not involve periodontally healthy participants (nine articles), did not use Salvadora persica products/rinses (38 articles), or did not include an applicable comparison group (eight articles). However, there were no studies that compared Salvadora persica with chlorhexidine as an antibacterial agent.

The search strategy was conducted according to the PRISMA 2009 guidelines,[40] and yielded 18 eligible articles (Figure 1). One other article was added from the reference lists of eligible articles. Thus, 19 articles were included in the qualitative analysis, all of which were on clinical trials discussing the antiplaque (15 articles), antibacterial (three articles), or both (one article) effects of Salvadora persica mouthrinses in comparison to either chlorhexidine or placebo. Non-randomized studies (one article) were excluded from the quantitative analysis to ensure the validity of the results.[28] Hence, 18 articles were included in the quantitative synthesis. Kappa score showed 100% perfect agreement between authors in the search strategy and which article to include or exclude from the analysis. We were able to obtain the mean and standard deviation (SD) values that were not available in one published study directly from the corresponding author.[32]

Figure 1.

Flow chart of study selection process

Additional file 3: Table S3 summarizes the characteristics of the articles that were included in the meta-analysis. The most common preparation used was a solution containing 7% Salvadora persica extract.[23,25,26,29,30,32–34,36,41] The studies were mostly performed in adult patients, up to 60 years old, whereas five studies were performed in children between the ages of 6 and 18 years.[20–22,27,29,36] Some studies were of cross-over nature, with a washout period ranging from 2 to 10 days[28,31,35] and 2 to 8 weeks.[9,23,29] The inclusion and exclusion criteria of the eligible studies are listed in Additional file 4: Table S4.

Meta-analysis Results

Studies measuring the ability of Salvadora persica to reduce the mean plaque scores were assessed. Figure 2a shows that there was a significant difference in the mean plaque scores of Salvadora persica as compared to chlorhexidine (P = 0.04, MD: 0.19, and 95% CI: 0.01 to 0.37). Furthermore, the post-intervention mean plaque scores of Salvadora persica (Figure 2b) showed statistically significant differences as compared to placebo (P < 0.00001, MD: -0.59, and 95% CI: -0.82 to − 0.35).

Figure 2.

2a. Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persica in comparison to both concentrations (0.12% & 0.2%) of chlorhexidine. 2b. Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persica in comparison to placebo. [SP Salvadora persica, CHXchlorhexidine, **: significant difference at p = 0.05]

Figure 32c shows the pre- and post-intervention values of all three mouthrinses: Salvadora persica, chlorhexidine, and placebo rinses demonstrated significant differences (P = 0.00001). Salvadora persica showed significant antiplaque effects as compared to the pre-intervention scores (P < 0.00001, MD: 0.46, and 95% CI: 0.29 to 0.63). Further, both 0.12 and 0.2% chlorhexidine significantly reduced plaque accumulation post-rinsing (P < 0.00001, MD: 1.79, and 95% CI: 1.39 to 2.19; and P < 0.0006, MD: 0.45, and 95% CI: 0.19 to 0.70, respectively). Placebo rinses showed nonsignificant differences between pre- and post-intervention (P = 0.07, MD: -0.19, and 95% CI: -0.39 to 0.02).

Figure 3.

2c: Forest plot for meta-analysis of the pre- and post-treatment mean plaque scores of: (2.1) Salvadora persica, (2.2) 0.12% chlorhexidine, (2.3) 0.2% chlorhexidine, and (2.4) placebo rinse. 2d: Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persica in comparison to 0.12% of chlorhexidine (sub-grouped according to concentration of chlorhexidine). 2e: Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persica in comparison to 0.2% of chlorhexidine (sub-grouped according to concentration of chlorhexidine). [SP Salvadora persica, CHX chlorhexidine, **: significant difference at p = 0.05]

Studies measuring the ability of Salvadora persica to reduce cariogenic bacterial loads were also assessed. The use of Salvadora persica-containing rinses resulted in significant decreases in both Streptococcus mutans (Figure 43a) and Lactobacillus counts (Figure 43d) as compared to the use of placebo rinses (P < 0.0001, MD: -1.42, and 95% CI: -2.08 to − 0.76; and P < 0.00001, MD: -1.12, and 95% CI: -1.45 to − 0.79, respectively).

Figure 4.

2h Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persica compared to chlorhexidine (sub-grouped according to refrain of oral mechanical measures); (2.1) studies that refrained mechanical measures and (2.2) studies that did not refrain such measures. 3a Forest plot for meta-analysis of post- treatment Streptococcus mutans counts of Salvadora persica in comparison to placebo. 3b Forest plot for meta-analysis of the pre- and post-treatment Streptococcus mutans counts of Salvadora persica. 3c Forest plot for meta-analysis of the pre- and post-treatment Streptococcus mutans counts of placebo rinses. 3d Forest plot for meta-analysis of post-treatment Lactobacilli counts of Salvadora persica in comparison to placebo. [SP Salvadora persica, CHX chlorhexidine, **: significant difference at p = 0.05]

Figure 43b demonstrates the post-treatment values for the anti-streptococcal effects of Salvadora persica and placebo. The use of Salvadora persica-containing rinses resulted in statistically significant differences in post-treatment scores as opposed to the pre-treatment values (P < 0.00001, MD: 1.82, and 95% CI: 1.03 to 2.62). It had similar effects in reducing lactobacillus counts (P < 0.00001, MD: 1.54, and 95% CI: 0.91 to 2.17), as shown in Figure 53e.

Figure 5.

3e Forest plot for meta-analysis of the pre- and post-treatment Lactobacilli counts of: (3.1) Salvadora persica and (3.2) placebo. **: significant difference at p=0.05. Funnel plot for meta-analysis of the pre- and post-treatment mean plaque scores of: Salvadora persica, 0.12 % chlorhexidine and 0.2% chlorhexidine

Assessing Risk of Bias

Although the quality of studies varied due to the use of different criteria, none were excluded based on quality level. Seven papers were regarded as high quality,[25,31,32,34,35,41] while the remaining papers were of moderate quality[9,20–24,26–30,36] . As for the risk of bias,

Figure 6 shows the author's judgments about each risk of bias item for each included study presented as percentages and as a summary. Additional file 2: Table S2 presents a more detailed appraisal of the risk of bias that includes the authors' assessment and comments.

Figure 6.

Risk of bias: review authors' judgements about each risk of bias item presented as percentages and as a summary

Since a non-randomized study was excluded from the quantitative analysis,[28] the grade of evidence was Ia with a strength of recommendation being class A for both the antiplaque and antibacterial effects, according to the system of Shekelle and colleagues.

Sensitivity Analysis

There was statistical heterogeneity in the pooled effect estimates. The subgrouping results according to the concentration of chlorhexidine used were similar to those prior to subgrouping, indicating that the result of the meta-analysis showed stability and reliability (Figure 32d, 32e). The results showed significant difference favoring the use of both 0.12 and 0.2% of chlorhexidine over Salvadora persica (P < 0.00001, MD: 0.52, and 95% CI: 0.35 to 0.69; and P = 0.010, MD 0.12, and 95% CI: 0.03 to 0.21, respectively). A similar conclusion is drawn when subgrouped according to the quality of the study (Figure 72f) in which high-quality studies favored the use of chlorhexidine (P = 0.003, MD: 0.19, and 95% CI: 0.06 to 0.31). When the data were subgrouped according to the duration of mouthwash use (Figure 72g), chlorhexidine exhibited superior effects when used for less than 3 weeks (P = 0.03, MD: 0.15, and 95% CI: 0.01 to 0.29), whereas its long-term use showed similar efficacy in relation to Salvadora persica in preventing plaque formation (P = 0.18, MD: 0.24, and 95% CI: -0.11 to 0.60). In Figure 42h, data are subgrouped according to whether the participants were asked to refrain from their daily oral mechanical measures of plaque control. Chlorhexidine demonstrated greater antiplaque effects in the absence of other plaque control measures (P = 0.01, MD: 0.22, and 95% CI: 0.05 to 0.39).

Figure 7.

2f Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persica compared to chlorhexidine (sub-grouped according to risk of bias assessment); (2.1) high quality studies (2.2) moderate quality studies. 2g Forest plot for meta-analysis of the post-intervention mean plaque scores of Salvadora persicacompared to chlorhexidine (subgrouped according to duration); (2.1) duration of < 3 weeks and (2.2) duration of ≥3 weeks. [SP Salvadora persica, CHX chlorhexidine, **: significant difference at p = 0.05]

However, subgrouping according to other variables and subgrouping studies assessing anticariogenic bacterial effects were not possible because of the limited number of studies. The funnel plot of the studies at pre- and post-intervention plaque scores of the three mouthrinses exhibited scattering indicating heterogenity (Figure 5).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....