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


BMC Oral Health. 2019;19(64) 

In This Article



This systematic review has been registered on PROSPERO (; register #CRD42018094678.

Search Strategy

This comprehensive systematic review included in-vivo studies reporting the use of Salvadora persica mouthrinse in comparison to chlorhexidine and/or placebo, in terms of its anticariogenic and/or antiplaque effects. The search strategy comprised keywords that were listed either separately or in combination with other words: ((Miswak) OR (herbal) OR (Siwak) OR (Arak) OR (peelu) OR (Salvadora persica)) AND ((chlorhexidine)) AND ((mouthwash) OR (mouthrinse)) AND ((antimicrobial) OR (antibacterial) OR (antiplaque)). All publications in MEDLINE-PubMed, Cochrane Central Register of Controlled Trials, Wiley Online Library, ScienceDirect, and Google Scholar were reviewed by two assessors (EJ and LB) independently. The search was done in December 2018 and sought to identify all articles related to the topic that had been published up to that date, with no constraints in terms of language. Moreover, the references of the identified studies were reviewed manually. Authors of four studies were contacted via email to acquire additional data that were not available in the published articles.

Screening Process

Two authors (EJ and LB) reviewed the titles and abstracts of all articles independently. Any disagreement among authors were settled in consensus meetings and in discussion with a third author (HS). After consensus, the three reviewers (EJ, LB, and HS) retrieved the full-text article for screening and data extraction. The references of the remaining articles were reviewed manually.

Eligibility Criteria

Papers were screened according to the following inclusion criteria:

  • Population (P): clinical trials and in-vivo studies conducted on medically healthy individuals;

  • Intervention (I): the use of Salvadora persica mouthwash;

  • Comparison (C): the use of chlorhexidine mouthwash and/or placebo; and

  • Outcome (O): a decrease in the mean plaque score and cariogenic bacterial counts.

The exclusion criteria were as follows:

  • Reviews of the literature, case–control, cross-sectional, and in-vitro studies;

  • Use of Salvadora persica sticks or toothpaste;

  • Use of mechanical methods for plaque removal;

  • Patients with periodontal breakdown and active caries;

  • Patients with orthodontic wires/brackets; and

  • Studies of the effects of Salvadora persica on other types of bacteria.

The following data were extracted by two of the authors (EJ and LB):

  • The study design and setting;

  • Total sample size, samples allocated per group, and sample description;

  • Eligibility criteria for including participants;

  • Baseline and post-treatment measurements of the intended outcomes in Salvadora persica mouthwash users versus other interventions;

  • The concentration of Salvadora persica used;

  • The frequency and duration of the interventions; and

  • The plaque index system used.

Risk of Bias Assessment

The quality of the studies that fulfilled the inclusion criteria were assessed according to the CONSORT 2010 checklist[37] used for randomized trials. The scale measured two items: methods and results, which included the trial design, participants, participant flow, interventions, outcomes, sample size, randomization, blinding, recruitment, baseline data, numbers analyzed, ancillary analyses, and harms, with a minimum score of 0 and a maximum score of 27. Studies with scores of 9 or less were regarded as being of low quality; 10–18 were considered to be of moderate quality; and studies with a score of 19 or more were considered as being of high quality. The details of this assessment is listed in Additional file 1: Table S1. As for the risk of bias, the included studies were assessed against the five main domains: selection bias, performance bias, attrition bias, reporting bias, and other sources of bias, in accordance with methods recommended by the Cochrane Handbook for Systematic Reviews of Interventions.[38] The following judgments were used: low risk, high risk, or unclear (either for the lack of information or uncertainty over the potential for bias). The details of this evaluation along with explanation and comments are listed in Additional file 2: Table S2. The quality assessment, as well as the risk of bias assessment, were evaluated by two review authors (EJ and LB) independently. Authors resolved any disagreement by consensus, and the third author (HS) was consulted to resolve any conflict if necessary.

Moreover, the strength of evidence and recommendation was evaluated using Shekelle and colleagues system.[39]

Statistical Analysis

The meta-analysis was carried out using the free Review Manager software (Cochrane Collaboration). A statistical test of homogeneity was implemented in order to determine whether the results of the separate studies could be combined. An inconsistency coefficient was calculated (I2 statistic), which was based on the chi-square test. If the value was 50% or more, it signified that the results of such studies were of moderate heterogeneity, whereas a value of 75% or more signified the presence of high heterogeneity between the studies.[38] Odd ratios with a fixed effect model was used for homogenous studies, whilst a random effect model was used for the heterogeneous studies. Forest plots were used to display mean differences (MD) and their 95% confidence interval (CI) of individual studies and a summary estimate of effect.

Sensitivity Analysis

Subgroup analyses were conducted to assess the stability of the studies. Papers that assessed the antiplaque effects of the mouthrinses were sub-grouped according to the concentration of chlorhexidine used, duration of mouthwash use, refraining of mechanical plaque measures, and strength of the included studies. The duration of mouthwash used was sub-divided into 3-week intervals. If the results of these analyses were similar to those performed prior to subgrouping, then it is unlikely to be the source of heterogeneity. Funnel plots were used to assess precision and heterogeneity as a consideration of possible small study effects.